Implement SBOM generation

Signed-off-by: Arkadii Yakovets <ark@victoriametrics.com>

.github/workflows/build.yml

@@ -0,0 +1,54 @@
+name: build
+
+on:
+  push:
+    branches:
+      - cluster
+      - master
+    paths:
+      - '**.go'
+      - '**/Dockerfile*' # The trailing * is for app/vmui/Dockerfile-*.
+      - '**/Makefile'
+  pull_request:
+    branches:
+      - cluster
+      - master
+    paths:
+      - '**.go'
+      - '**/Dockerfile*' # The trailing * is for app/vmui/Dockerfile-*.
+      - '**/Makefile'
+
+permissions:
+  contents: read
+
+concurrency:
+  cancel-in-progress: true
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+
+jobs:
+  build:
+    name: Build
+    runs-on: ubuntu-latest
+    steps:
+      - name: Code checkout
+        uses: actions/checkout@v4
+
+      - name: Setup Go
+        id: go
+        uses: actions/setup-go@v5
+        with:
+          go-version: stable
+          cache: false
+
+      - name: Cache Go artifacts
+        uses: actions/cache@v4
+        with:
+          path: |
+            ~/.cache/go-build
+            ~/go/bin
+            ~/go/pkg/mod
+          key: go-artifacts-${{ runner.os }}-crossbuild-${{ steps.go.outputs.go-version }}-${{ hashFiles('go.sum', 'Makefile', 'app/**/Makefile') }}
+          restore-keys: go-artifacts-${{ runner.os }}-crossbuild-
+
+      - name: Run crossbuild
+        run: make crossbuild

.github/workflows/codeql-analysis-go.yml

@@ -0,0 +1,62 @@
+name: 'CodeQL Go'
+
+on:
+  push:
+    branches:
+      - cluster
+      - master
+    paths:
+      - '**.go'
+  pull_request:
+    branches:
+      - cluster
+      - master
+    paths:
+      - '**.go'
+
+concurrency:
+  cancel-in-progress: true
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+
+jobs:
+  analyze:
+    name: Analyze
+    runs-on: ubuntu-latest
+    permissions:
+      actions: read
+      contents: read
+      security-events: write
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Set up Go
+        id: go
+        uses: actions/setup-go@v5
+        with:
+          cache: false
+          go-version: stable
+
+      - name: Cache Go artifacts
+        uses: actions/cache@v4
+        with:
+          path: |
+            ~/.cache/go-build
+            ~/go/bin
+            ~/go/pkg/mod
+          key: go-artifacts-${{ runner.os }}-codeql-analyze-${{ steps.go.outputs.go-version }}-${{ hashFiles('go.sum', 'Makefile', 'app/**/Makefile') }}
+          restore-keys: go-artifacts-${{ runner.os }}-codeql-analyze-
+
+      - name: Initialize CodeQL
+        uses: github/codeql-action/init@v3
+        with:
+          languages: go
+
+      - name: Autobuild
+        uses: github/codeql-action/autobuild@v3
+
+      - name: Perform CodeQL Analysis
+        uses: github/codeql-action/analyze@v3
+        with:
+          category: 'language:go'

.github/workflows/codeql-analysis-js-ts.yml

@@ -1,22 +1,27 @@
-name: "CodeQL - JS"
+name: 'CodeQL JS/TS'
 
 on:
   push:
-    branches: [master, cluster]
+    branches:
+      - cluster
+      - master
     paths:
-      - "**.js"
+      - '**.js'
+      - '**.ts'
+      - '**.tsx'
   pull_request:
-    # The branches below must be a subset of the branches above
-    branches: [master, cluster]
+    branches:
+      - cluster
+      - master
     paths:
-      - "**.js"
-  schedule:
-    - cron: "30 18 * * 2"
+      - '**.js'
+      - '**.ts'
+      - '**.tsx'
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
   cancel-in-progress: true
-  
+
 jobs:
   analyze:
     name: Analyze
@@ -26,11 +31,6 @@ jobs:
       contents: read
       security-events: write
 
-    strategy:
-      fail-fast: false
-      matrix:
-        language: ["javascript"]
-
     steps:
       - name: Checkout repository
         uses: actions/checkout@v4
@@ -38,9 +38,9 @@ jobs:
       - name: Initialize CodeQL
         uses: github/codeql-action/init@v3
         with:
-          languages: ${{ matrix.language }}
+          languages: javascript-typescript
 
       - name: Perform CodeQL Analysis
         uses: github/codeql-action/analyze@v3
         with:
-          category: "javascript"
+          category: 'language:js/ts'

.github/workflows/codeql-analysis.yml

@@ -1,103 +0,0 @@
-# For most projects, this workflow file will not need changing; you simply need
-# to commit it to your repository.
-#
-# You may wish to alter this file to override the set of languages analyzed,
-# or to provide custom queries or build logic.
-#
-# ******** NOTE ********
-# We have attempted to detect the languages in your repository. Please check
-# the `language` matrix defined below to confirm you have the correct set of
-# supported CodeQL languages.
-#
-name: "CodeQL"
-
-on:
-  push:
-    branches: [master, cluster]
-    paths-ignore:
-      - "docs/**"
-      - "**.md"
-      - "**.txt"
-      - "**.js"
-  pull_request:
-    # The branches below must be a subset of the branches above
-    branches: [master, cluster]
-    paths-ignore:
-      - "docs/**"
-      - "**.md"
-      - "**.txt"
-      - "**.js"
-  schedule:
-    - cron: "30 18 * * 2"
-
-concurrency:
-  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
-  cancel-in-progress: true
-
-jobs:
-  analyze:
-    name: Analyze
-    runs-on: ubuntu-latest
-    permissions:
-      actions: read
-      contents: read
-      security-events: write
-
-    strategy:
-      fail-fast: false
-      matrix:
-        language: ["go"]
-        # CodeQL supports [ 'cpp', 'csharp', 'go', 'java', 'javascript', 'python', 'ruby' ]
-        # Learn more about CodeQL language support at https://git.io/codeql-language-support
-
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@v4
-
-      - name: Set up Go
-        id: go
-        uses: actions/setup-go@v5
-        with:
-          go-version: stable
-          cache: false
-        if: ${{ matrix.language == 'go' }}
-
-      - name: Cache Go artifacts
-        uses: actions/cache@v4
-        with:
-          path: |
-            ~/.cache/go-build
-            ~/go/pkg/mod
-            ~/go/bin
-          key: go-artifacts-${{ runner.os }}-codeql-analyze-${{ steps.go.outputs.go-version }}-${{ hashFiles('go.sum', 'Makefile', 'app/**/Makefile') }}
-          restore-keys: go-artifacts-${{ runner.os }}-codeql-analyze-
-        if: ${{ matrix.language == 'go' }}
-
-      # Initializes the CodeQL tools for scanning.
-      - name: Initialize CodeQL
-        uses: github/codeql-action/init@v3
-        with:
-          languages: ${{ matrix.language }}
-          # If you wish to specify custom queries, you can do so here or in a config file.
-          # By default, queries listed here will override any specified in a config file.
-          # Prefix the list here with "+" to use these queries and those in the config file.
-          # queries: ./path/to/local/query, your-org/your-repo/queries@main
-
-      # Autobuild attempts to build any compiled languages  (C/C++, C#, or Java).
-      # If this step fails, then you should remove it and run the build manually (see below)
-      - name: Autobuild
-        uses: github/codeql-action/autobuild@v3
-
-      # ℹ️ Command-line programs to run using the OS shell.
-      # 📚 https://git.io/JvXDl
-
-      # ✏️ If the Autobuild fails above, remove it and uncomment the following three lines
-      #    and modify them (or add more) to build your code if your project
-      #    uses a compiled language
-
-      #- run: |
-      #   make bootstrap
-      #   make release
-
-      - name: Perform CodeQL Analysis
-        uses: github/codeql-action/analyze@v3

.github/workflows/main.yml

@@ -1,29 +1,25 @@
 name: main
+
 on:
   push:
     branches:
-      - master
       - cluster
-    paths-ignore:
-      - "docs/**"
-      - "**.md"
-      - "dashboards/**"
-      - "deployment/**.yml"
+      - master
+    paths:
+      - '**.go'
   pull_request:
     branches:
-      - master
       - cluster
-    paths-ignore:
-      - "docs/**"
-      - "**.md"
-      - "dashboards/**"
-      - "deployment/**.yml"
+      - master
+    paths:
+      - '**.go'
+
 permissions:
   contents: read
 
 concurrency:
-  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
   cancel-in-progress: true
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
 
 jobs:
   lint:
@@ -37,16 +33,16 @@ jobs:
         id: go
         uses: actions/setup-go@v5
         with:
-          go-version: stable
           cache: false
+          go-version: stable
 
       - name: Cache Go artifacts
         uses: actions/cache@v4
         with:
           path: |
             ~/.cache/go-build
-            ~/go/pkg/mod
             ~/go/bin
+            ~/go/pkg/mod
           key: go-artifacts-${{ runner.os }}-check-all-${{ steps.go.outputs.go-version }}-${{ hashFiles('go.sum', 'Makefile', 'app/**/Makefile') }}
           restore-keys: go-artifacts-${{ runner.os }}-check-all-
 
@@ -55,41 +51,18 @@ jobs:
           make check-all
           git diff --exit-code
 
-  build:
-    needs: lint
-    name: build
-    runs-on: ubuntu-latest
-    steps:
-      - name: Code checkout
-        uses: actions/checkout@v4
-
-      - name: Setup Go
-        id: go
-        uses: actions/setup-go@v5
-        with:
-          go-version: stable
-          cache: false
-
-      - name: Cache Go artifacts
-        uses: actions/cache@v4
-        with:
-          path: |
-            ~/.cache/go-build
-            ~/go/pkg/mod
-            ~/go/bin
-          key: go-artifacts-${{ runner.os }}-crossbuild-${{ steps.go.outputs.go-version }}-${{ hashFiles('go.sum', 'Makefile', 'app/**/Makefile') }}
-          restore-keys: go-artifacts-${{ runner.os }}-crossbuild-
-
-      - name: Build
-        run: make crossbuild
-
   test:
+    name: test
     needs: lint
+    runs-on: ubuntu-latest
+
     strategy:
       matrix:
-        scenario: ["test-full", "test-pure", "test-full-386"]
-    name: test
-    runs-on: ubuntu-latest
+        scenario:
+          - 'test-full'
+          - 'test-full-386'
+          - 'test-pure'
+
     steps:
       - name: Code checkout
         uses: actions/checkout@v4
@@ -98,20 +71,20 @@ jobs:
         id: go
         uses: actions/setup-go@v5
         with:
-          go-version: stable
           cache: false
+          go-version: stable
 
       - name: Cache Go artifacts
         uses: actions/cache@v4
         with:
           path: |
             ~/.cache/go-build
-            ~/go/pkg/mod
             ~/go/bin
+            ~/go/pkg/mod
           key: go-artifacts-${{ runner.os }}-${{ matrix.scenario }}-${{ steps.go.outputs.go-version }}-${{ hashFiles('go.sum', 'Makefile', 'app/**/Makefile') }}
           restore-keys: go-artifacts-${{ runner.os }}-${{ matrix.scenario }}-
 
-      - name: run tests
+      - name: Run tests
         run: make ${{ matrix.scenario}}
 
       - name: Publish coverage

.github/workflows/sync-docs.yml

@@ -25,7 +25,7 @@ jobs:
           token: ${{ secrets.VM_BOT_GH_TOKEN }}
           path: docs
       - name: Import GPG key
-        uses: crazy-max/ghaction-import-gpg@v5
+        uses: crazy-max/ghaction-import-gpg@v6
         with:
           gpg_private_key: ${{ secrets.VM_BOT_GPG_PRIVATE_KEY }}
           passphrase: ${{ secrets.VM_BOT_PASSPHRASE }}

.github/workflows/test_package.yml

@@ -0,0 +1,59 @@
+name: Package tests
+
+on:
+  push:
+    branches:
+      - cluster
+      - master
+    paths:
+      - '.github/workflows/test_package.yml'
+      - '**/Dockerfile*'
+      - '**/Makefile'
+  pull_request:
+    branches:
+      - cluster
+      - master
+    paths:
+      - '.github/workflows/test_package.yml'
+      - '**/Dockerfile*'
+      - '**/Makefile'
+
+permissions:
+  contents: read
+
+concurrency:
+  cancel-in-progress: true
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+
+jobs:
+  test_release:
+    name: Test Release
+    runs-on: ubuntu-latest
+    steps:
+      - name: Code checkout
+        uses: actions/checkout@v4
+
+      - name: Setup Go
+        id: go
+        uses: actions/setup-go@v5
+        with:
+          go-version: stable
+          cache: false
+
+      - name: Cache Go artifacts
+        uses: actions/cache@v4
+        with:
+          path: |
+            ~/.cache/go-build
+            ~/go/bin
+            ~/go/pkg/mod
+          key: go-artifacts-${{ runner.os }}-test-release-${{ steps.go.outputs.go-version }}-${{ hashFiles('Makefile', '**/Dockerfile*', '**/Makefile') }}
+          restore-keys: go-artifacts-${{ runner.os }}-test-release-
+
+      - name: Run make release
+        run: |
+          make clean
+          make release
+
+      - name: Run release tests
+        run: make test-release-victoria-metrics

.gitignore

@@ -22,4 +22,6 @@ Gemfile.lock
 /_site
 _site
 *.tmp
-/docs/.jekyll-metadata
+/docs/.jekyll-metadata
+coverage.txt
+cspell.json

CODE_OF_CONDUCT_RU.md

@@ -1,120 +0,0 @@
-
-# Кодекс Поведения участника
-
-## Наши обязательства
-
-Мы, как участники, авторы и лидеры обязуемся сделать участие в сообществе
-свободным от притеснений для всех, независимо от возраста, телосложения,
-видимых или невидимых ограничений способности, этнической принадлежности,
-половых признаков, гендерной идентичности и выражения, уровня опыта,
-образования, социо-экономического статуса, национальности, внешности,
-расы, религии, или сексуальной идентичности и ориентации.
-
-Мы обещаем действовать и взаимодействовать таким образом, чтобы вносить вклад в открытое,
-дружелюбное, многообразное, инклюзивное и здоровое сообщество.
-
-## Наши стандарты
-
-Примеры поведения, создающие условия для благоприятных взаимоотношений включают в себя:
-
-* Проявление доброты и эмпатии к другим участникам проекта
-* Уважение к чужой точке зрения и опыту
-* Конструктивная критика и принятие конструктивной критики
-* Принятие ответственности, принесение извинений тем, кто пострадал от наших ошибок
-  и извлечение уроков из опыта
-* Ориентирование на то, что лучше подходит для сообщества, а не только для нас лично
-
-Примеры неприемлемого поведения участников включают в себя:
-
-* Использование выражений или изображений сексуального характера и нежелательное сексуальное внимание или домогательство в любой форме
-* Троллинг, оскорбительные или уничижительные комментарии, переход на личности или затрагивание политических убеждений
-* Публичное или приватное домогательство
-* Публикация личной информации других лиц, например, физического или электронного адреса, без явного разрешения
-* Иное поведение, которое обоснованно считать неуместным в профессиональной обстановке
-
-## Обязанности
-
-Лидеры сообщества отвечают за разъяснение и применение наших стандартов приемлемого
-поведения и будут предпринимать соответствующие и честные меры по исправлению положения
-в ответ на любое поведение, которое они сочтут неприемлемым, угрожающим, оскорбительным или вредным.
-
-Лидеры сообщества обладают правом и обязанностью удалять, редактировать или отклонять
-комментарии, коммиты, код, изменения в вики, вопросы и другой вклад, который не совпадает
-с Кодексом Поведения, и предоставят причины принятого решения, когда сочтут нужным.
-
-## Область применения
-
-Данный Кодекс Поведения применим во всех во всех публичных физических и цифровых пространства сообщества,
-а также когда человек официально представляет сообщество в публичных местах.
-Примеры представления проекта или сообщества включают использование официальной электронной почты,
-публикации в официальном аккаунте в социальных сетях,
-или упоминания как представителя в онлайн или оффлайн мероприятии.
-
-## Приведение в исполнение
-
-О случаях домогательства, а так же оскорбительного или иного другого неприемлемого
-поведения можно сообщить ответственным лидерам сообщества с помощью письма на info@victoriametrics.com
-Все жалобы будут рассмотрены и расследованы оперативно и беспристрастно.
-
-Все лидеры сообщества обязаны уважать неприкосновенность частной жизни и личную
-неприкосновенность автора сообщения.
-
-## Руководство по исполнению
-
-Лидеры сообщества будут следовать следующим Принципам Воздействия в Сообществе,
-чтобы определить последствия для тех, кого они считают виновными в нарушении данного Кодекса Поведения:
-
-### 1. Исправление
-
-**Общественное влияние**: Использование недопустимой лексики или другое поведение,
-считающиеся непрофессиональным или нежелательным в сообществе.
-
-**Последствия**: Личное, письменное предупреждение от лидеров сообщества,
-объясняющее суть нарушения и почему такое поведение
-было неуместно. Лидеры сообщества могут попросить принести публичное извинение.
-
-### 2. Предупреждение
-
-**Общественное влияние**: Нарушение в результате одного инцидента или серии действий.
-
-**Последствия**: Предупреждение о последствиях в случае продолжающегося неуместного поведения.
-На определенное время не допускается взаимодействие с людьми, вовлеченными в инцидент,
-включая незапрошенное взаимодействие
-с теми, кто обеспечивает соблюдение Кодекса. Это включает в себя избегание взаимодействия
-в публичных пространствах, а так же во внешних каналах,
-таких как социальные сети. Нарушение этих правил влечет за собой временный или вечный бан.
-
-### 3. Временный бан
-
-**Общественное влияние**: Серьёзное нарушение стандартов сообщества,
-включая продолжительное неуместное поведение.
-
-**Последствия**: Временный запрет (бан) на любое взаимодействие
-или публичное общение с сообществом на определенный период времени.
-На этот период не допускается публичное или личное взаимодействие с людьми,
-вовлеченными в инцидент, включая незапрошенное взаимодействие
-с теми, кто обеспечивает соблюдение Кодекса.
-Нарушение этих правил влечет за собой вечный бан.
-
-### 4. Вечный бан
-
-**Общественное влияние**: Демонстрация систематических нарушений стандартов сообщества,
-включая продолжающееся неуместное поведение, домогательство до отдельных лиц,
-или проявление агрессии либо пренебрежительного отношения к категориям лиц.
-
-**Последствия**: Вечный запрет на любое публичное взаимодействие с сообществом.
-
-## Атрибуция
-
-Данный Кодекс Поведения основан на [Кодекс Поведения участника][homepage],
-версии 2.0, доступной по адресу
-<https://www.contributor-covenant.org/version/2/0/code_of_conduct.html>.
-
-Принципы Воздействия в Сообществе были вдохновлены [Mozilla's code of conduct
-enforcement ladder](https://github.com/mozilla/diversity).
-
-[homepage]: https://www.contributor-covenant.org
-
-Ответы на общие вопросы о данном кодексе поведения ищите на странице FAQ:
-<https://www.contributor-covenant.org/faq>. Переводы доступны по адресу
-<https://www.contributor-covenant.org/translations>.

Makefile

@@ -14,15 +14,86 @@ endif
 
 GO_BUILDINFO = -X '$(PKG_PREFIX)/lib/buildinfo.Version=$(APP_NAME)-$(DATEINFO_TAG)-$(BUILDINFO_TAG)'
 
+VICTORIA_LOGS_COMPONENTS = victoria-logs
+VICTORIA_METRICS_COMPONENTS = victoria-metrics
+VICTORIA_METRICS_UTILS_COMPONENTS = vmagent vmalert vmalert-tool vmauth vmbackup vmrestore vmctl
+
 .PHONY: $(MAKECMDGOALS)
 
 include app/*/Makefile
+include cspell/Makefile
 include docs/Makefile
 include deployment/*/Makefile
 include dashboards/Makefile
-include snap/local/Makefile
 include package/release/Makefile
 
+define RELEASE_GOOS_GOARCH
+	$(eval PKG_NAME := $(1))
+	$(eval PKG_COMPONENTS := $(2))
+
+	# Build
+	$(foreach pkg_component, $(PKG_COMPONENTS), $(MAKE) $(pkg_component)-$(GOOS)-$(GOARCH)-prod)
+
+	# Generate SBOM
+	mkdir -p "bin/$(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom"
+	$(foreach pkg_component, $(PKG_COMPONENTS),
+		cyclonedx-gomod app -assert-licenses -json -licenses -packages \
+			-main app/$(pkg_component) \
+			-output bin/$(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom/$(pkg_component)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom.json)
+	
+	# Pack and compress
+	$(foreach pkg_component, $(PKG_COMPONENTS),
+		cd bin && tar --transform="flags=r;s|-$(GOOS)-$(GOARCH)||" -rf $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar \
+			$(pkg_component)-$(GOOS)-$(GOARCH)-prod)
+	cd bin && gzip $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar
+	cd bin && tar -czf $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom.tar.gz $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom
+
+	# Generate checksums
+	cd bin && \
+		sha256sum $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz > $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_checksums.txt && \
+		sha256sum $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom.tar.gz >> $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_checksums.txt
+	$(foreach pkg_component, $(PKG_COMPONENTS),
+		cd bin && sha256sum $(pkg_component)-$(GOOS)-$(GOARCH)-prod | sed s/-$(GOOS)-$(GOARCH)-prod/-prod/ >> \
+			$(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_checksums.txt)
+
+	# Clean up
+	cd bin && rm -rf $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom
+	$(foreach pkg_component, $(PKG_COMPONENTS), cd bin && rm -f $(pkg_component)-$(GOOS)-$(GOARCH)-prod)
+endef
+
+define RELEASE_WINDOWS_GOARCH
+	$(eval PKG_NAME := $(1))
+	$(eval PKG_COMPONENTS := $(2))
+
+	# Build
+	$(foreach pkg_component, $(PKG_COMPONENTS), $(MAKE) $(pkg_component)-$(GOOS)-$(GOARCH)-prod)
+
+	# Generate SBOM
+	mkdir -p "bin/$(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom"
+	$(foreach pkg_component, $(PKG_COMPONENTS),
+		cyclonedx-gomod app -assert-licenses -json -licenses -packages \
+			-main app/$(pkg_component) \
+			-output bin/$(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom/$(pkg_component)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom.json)
+	
+	# Pack and compress
+	$(foreach pkg_component, $(PKG_COMPONENTS),
+		cd bin && zip -u $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG).zip \
+			$(pkg_component)-$(GOOS)-$(GOARCH)-prod.exe)
+	cd bin && zip $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom.zip -r $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom
+
+	# Generate checksums
+	cd bin && \
+		sha256sum $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG).zip > $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_checksums.txt && \
+		sha256sum $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom.zip >> $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_checksums.txt
+	$(foreach pkg_component, $(PKG_COMPONENTS),
+		cd bin && sha256sum $(pkg_component)-$(GOOS)-$(GOARCH)-prod.exe >> $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_checksums.txt)
+
+	# Clean up
+	cd bin && rm -rf $(PKG_NAME)-$(GOOS)-$(GOARCH)-$(PKG_TAG)_bom
+	$(foreach pkg_component, $(PKG_COMPONENTS), cd bin && rm -f $(pkg_component)-$(GOOS)-$(GOARCH)-prod.exe)
+endef
+
+
 all: \
 	victoria-metrics-prod \
 	victoria-logs-prod \
@@ -241,26 +312,13 @@ release-victoria-metrics-openbsd-amd64:
 	GOOS=openbsd GOARCH=amd64 $(MAKE) release-victoria-metrics-goos-goarch
 
 release-victoria-metrics-windows-amd64:
-	GOARCH=amd64 $(MAKE) release-victoria-metrics-windows-goarch
+	GOOS=windows GOARCH=amd64 $(MAKE) release-victoria-metrics-windows-goarch
 
-release-victoria-metrics-goos-goarch: victoria-metrics-$(GOOS)-$(GOARCH)-prod
-	cd bin && \
-		tar --transform="flags=r;s|-$(GOOS)-$(GOARCH)||" -czf victoria-metrics-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
-			victoria-metrics-$(GOOS)-$(GOARCH)-prod \
-		&& sha256sum victoria-metrics-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
-			victoria-metrics-$(GOOS)-$(GOARCH)-prod \
-			| sed s/-$(GOOS)-$(GOARCH)-prod/-prod/ > victoria-metrics-$(GOOS)-$(GOARCH)-$(PKG_TAG)_checksums.txt
-	cd bin && rm -rf victoria-metrics-$(GOOS)-$(GOARCH)-prod
+release-victoria-metrics-goos-goarch:
+	$(call RELEASE_GOOS_GOARCH,victoria-metrics,$(VICTORIA_METRICS_COMPONENTS))
 
-release-victoria-metrics-windows-goarch: victoria-metrics-windows-$(GOARCH)-prod
-	cd bin && \
-		zip victoria-metrics-windows-$(GOARCH)-$(PKG_TAG).zip \
-			victoria-metrics-windows-$(GOARCH)-prod.exe \
-		&& sha256sum victoria-metrics-windows-$(GOARCH)-$(PKG_TAG).zip \
-			victoria-metrics-windows-$(GOARCH)-prod.exe \
-			> victoria-metrics-windows-$(GOARCH)-$(PKG_TAG)_checksums.txt
-	cd bin && rm -rf \
-		victoria-metrics-windows-$(GOARCH)-prod.exe
+release-victoria-metrics-windows-goarch:
+	$(call RELEASE_WINDOWS_GOARCH,victoria-metrics,$(VICTORIA_METRICS_COMPONENTS))
 
 release-victoria-logs:
 	$(MAKE_PARALLEL) release-victoria-logs-linux-386 \
@@ -355,77 +413,13 @@ release-vmutils-openbsd-amd64:
 	GOOS=openbsd GOARCH=amd64 $(MAKE) release-vmutils-goos-goarch
 
 release-vmutils-windows-amd64:
-	GOARCH=amd64 $(MAKE) release-vmutils-windows-goarch
+	GOOS=windows GOARCH=amd64 $(MAKE) release-vmutils-windows-goarch
 
-release-vmutils-goos-goarch: \
-	vmagent-$(GOOS)-$(GOARCH)-prod \
-	vmalert-$(GOOS)-$(GOARCH)-prod \
-	vmalert-tool-$(GOOS)-$(GOARCH)-prod \
-	vmauth-$(GOOS)-$(GOARCH)-prod \
-	vmbackup-$(GOOS)-$(GOARCH)-prod \
-	vmrestore-$(GOOS)-$(GOARCH)-prod \
-	vmctl-$(GOOS)-$(GOARCH)-prod
-	cd bin && \
-		tar --transform="flags=r;s|-$(GOOS)-$(GOARCH)||" -czf vmutils-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
-			vmagent-$(GOOS)-$(GOARCH)-prod \
-			vmalert-$(GOOS)-$(GOARCH)-prod \
-			vmalert-tool-$(GOOS)-$(GOARCH)-prod \
-			vmauth-$(GOOS)-$(GOARCH)-prod \
-			vmbackup-$(GOOS)-$(GOARCH)-prod \
-			vmrestore-$(GOOS)-$(GOARCH)-prod \
-			vmctl-$(GOOS)-$(GOARCH)-prod \
-		&& sha256sum vmutils-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
-			vmagent-$(GOOS)-$(GOARCH)-prod \
-			vmalert-$(GOOS)-$(GOARCH)-prod \
-			vmalert-tool-$(GOOS)-$(GOARCH)-prod \
-			vmauth-$(GOOS)-$(GOARCH)-prod \
-			vmbackup-$(GOOS)-$(GOARCH)-prod \
-			vmrestore-$(GOOS)-$(GOARCH)-prod \
-			vmctl-$(GOOS)-$(GOARCH)-prod \
-			| sed s/-$(GOOS)-$(GOARCH)-prod/-prod/ > vmutils-$(GOOS)-$(GOARCH)-$(PKG_TAG)_checksums.txt
-	cd bin && rm -rf \
-		vmagent-$(GOOS)-$(GOARCH)-prod \
-		vmalert-$(GOOS)-$(GOARCH)-prod \
-		vmalert-tool-$(GOOS)-$(GOARCH)-prod \
-		vmauth-$(GOOS)-$(GOARCH)-prod \
-		vmbackup-$(GOOS)-$(GOARCH)-prod \
-		vmrestore-$(GOOS)-$(GOARCH)-prod \
-		vmctl-$(GOOS)-$(GOARCH)-prod
+release-vmutils-goos-goarch:
+	$(call RELEASE_GOOS_GOARCH, vmutils, $(VICTORIA_METRICS_UTILS_COMPONENTS))
 
-release-vmutils-windows-goarch: \
-	vmagent-windows-$(GOARCH)-prod \
-	vmalert-windows-$(GOARCH)-prod \
-	vmalert-tool-windows-$(GOARCH)-prod \
-	vmauth-windows-$(GOARCH)-prod \
-	vmbackup-windows-$(GOARCH)-prod \
-	vmrestore-windows-$(GOARCH)-prod \
-	vmctl-windows-$(GOARCH)-prod
-	cd bin && \
-		zip vmutils-windows-$(GOARCH)-$(PKG_TAG).zip \
-			vmagent-windows-$(GOARCH)-prod.exe \
-			vmalert-windows-$(GOARCH)-prod.exe \
-			vmalert-tool-windows-$(GOARCH)-prod.exe \
-			vmauth-windows-$(GOARCH)-prod.exe \
-			vmbackup-windows-$(GOARCH)-prod.exe \
-			vmrestore-windows-$(GOARCH)-prod.exe \
-			vmctl-windows-$(GOARCH)-prod.exe \
-		&& sha256sum vmutils-windows-$(GOARCH)-$(PKG_TAG).zip \
-			vmagent-windows-$(GOARCH)-prod.exe \
-			vmalert-windows-$(GOARCH)-prod.exe \
-			vmalert-tool-windows-$(GOARCH)-prod.exe \
-			vmauth-windows-$(GOARCH)-prod.exe \
-			vmbackup-windows-$(GOARCH)-prod.exe \
-			vmrestore-windows-$(GOARCH)-prod.exe \
-			vmctl-windows-$(GOARCH)-prod.exe \
-			> vmutils-windows-$(GOARCH)-$(PKG_TAG)_checksums.txt
-	cd bin && rm -rf \
-		vmagent-windows-$(GOARCH)-prod.exe \
-		vmalert-windows-$(GOARCH)-prod.exe \
-		vmalert-tool-windows-$(GOARCH)-prod.exe \
-		vmauth-windows-$(GOARCH)-prod.exe \
-		vmbackup-windows-$(GOARCH)-prod.exe \
-		vmrestore-windows-$(GOARCH)-prod.exe \
-		vmctl-windows-$(GOARCH)-prod.exe
+release-vmutils-windows-goarch:
+	$(call RELEASE_WINDOWS_GOARCH, vmutils, $(VICTORIA_METRICS_UTILS_COMPONENTS))
 
 pprof-cpu:
 	go tool pprof -trim_path=github.com/VictoriaMetrics/VictoriaMetrics@ $(PPROF_FILE)
@@ -440,6 +434,8 @@ vet:
 
 check-all: fmt vet golangci-lint govulncheck
 
+clean-checkers: remove-golangci-lint remove-govulncheck
+
 test:
 	go test ./lib/... ./app/...
 
@@ -494,18 +490,27 @@ golangci-lint: install-golangci-lint
 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
 
+remove-golangci-lint:
+	rm -rf `which golangci-lint`
+
 govulncheck: install-govulncheck
 	govulncheck ./...
 
 install-govulncheck:
 	which govulncheck || go install golang.org/x/vuln/cmd/govulncheck@latest
 
+remove-govulncheck:
+	rm -rf `which govulncheck`
+
 install-wwhrd:
 	which wwhrd || go install github.com/frapposelli/wwhrd@latest
 
 check-licenses: install-wwhrd
 	wwhrd check -f .wwhrd.yml
 
+cyclonedx-gomod-install:
+	which cyclonedx-gomod || go install github.com/CycloneDX/cyclonedx-gomod/cmd/cyclonedx-gomod@latest
+
 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
@@ -523,6 +528,7 @@ copy-docs:
 	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.

README.md

@@ -2,21 +2,23 @@
 
 [![Latest Release](https://img.shields.io/github/release/VictoriaMetrics/VictoriaMetrics.svg?style=flat-square)](https://github.com/VictoriaMetrics/VictoriaMetrics/releases/latest)
 [![Docker Pulls](https://img.shields.io/docker/pulls/victoriametrics/victoria-metrics.svg?maxAge=604800)](https://hub.docker.com/r/victoriametrics/victoria-metrics)
-[![victoriametrics](https://snapcraft.io/victoriametrics/badge.svg)](https://snapcraft.io/victoriametrics)
 [![Slack](https://img.shields.io/badge/join%20slack-%23victoriametrics-brightgreen.svg)](https://slack.victoriametrics.com/)
 [![GitHub license](https://img.shields.io/github/license/VictoriaMetrics/VictoriaMetrics.svg)](https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/LICENSE)
 [![Go Report](https://goreportcard.com/badge/github.com/VictoriaMetrics/VictoriaMetrics)](https://goreportcard.com/report/github.com/VictoriaMetrics/VictoriaMetrics)
 [![Build Status](https://github.com/VictoriaMetrics/VictoriaMetrics/workflows/main/badge.svg)](https://github.com/VictoriaMetrics/VictoriaMetrics/actions)
 [![codecov](https://codecov.io/gh/VictoriaMetrics/VictoriaMetrics/branch/master/graph/badge.svg)](https://codecov.io/gh/VictoriaMetrics/VictoriaMetrics)
 
-<img src="docs/logo.webp" width="300" alt="VictoriaMetrics logo">
+<picture>
+  <source srcset="docs/logo_white.webp" media="(prefers-color-scheme: dark)">
+  <source srcset="docs/logo.webp" media="(prefers-color-scheme: light)">
+  <img src="docs/logo.webp" width="300" alt="VictoriaMetrics logo">
+</picture>
 
 VictoriaMetrics is a fast, cost-effective and scalable monitoring solution and time series database.
 See [case studies for VictoriaMetrics](https://docs.victoriametrics.com/casestudies/).
 
 VictoriaMetrics is available in [binary releases](https://github.com/VictoriaMetrics/VictoriaMetrics/releases/latest),
-[Docker images](https://hub.docker.com/r/victoriametrics/victoria-metrics/), [Snap packages](https://snapcraft.io/victoriametrics)
-and [source code](https://github.com/VictoriaMetrics/VictoriaMetrics). 
+[Docker images](https://hub.docker.com/r/victoriametrics/victoria-metrics/) and [source code](https://github.com/VictoriaMetrics/VictoriaMetrics). 
 
 Documentation for the cluster version of VictoriaMetrics is available [here](https://docs.victoriametrics.com/cluster-victoriametrics/).
 
@@ -86,7 +88,6 @@ VictoriaMetrics has the following prominent features:
   * [Prometheus exposition format](#how-to-import-data-in-prometheus-exposition-format).
   * [InfluxDB line protocol](#how-to-send-data-from-influxdb-compatible-agents-such-as-telegraf) over HTTP, TCP and UDP.
   * [Graphite plaintext protocol](#how-to-send-data-from-graphite-compatible-agents-such-as-statsd) with [tags](https://graphite.readthedocs.io/en/latest/tags.html#carbon).
-  * [Statsd plaintext protocol](#how-to-send-data-from-statsd-compatible-clients)
   * [OpenTSDB put message](#sending-data-via-telnet-put-protocol).
   * [HTTP OpenTSDB /api/put requests](#sending-opentsdb-data-via-http-apiput-requests).
   * [JSON line format](#how-to-import-data-in-json-line-format).
@@ -139,7 +140,6 @@ VictoriaMetrics can also be installed via these installation methods:
 * [Ansible role for installing cluster VictoriaMetrics (by VictoriaMetrics)](https://github.com/VictoriaMetrics/ansible-playbooks).
 * [Ansible role for installing cluster VictoriaMetrics (by community)](https://github.com/Slapper/ansible-victoriametrics-cluster-role).
 * [Ansible role for installing single-node VictoriaMetrics (by community)](https://github.com/dreamteam-gg/ansible-victoriametrics-role).
-* [Snap package for VictoriaMetrics](https://snapcraft.io/victoriametrics).
 
 ### How to start VictoriaMetrics
 
@@ -180,27 +180,6 @@ Additionally, all the VictoriaMetrics components allow setting flag values via e
 * For repeating flags an alternative syntax can be used by joining the different values into one using `,` char as separator (for example `-storageNode <nodeA> -storageNode <nodeB>` will translate to `storageNode=<nodeA>,<nodeB>`).
 * Environment var prefix can be set via `-envflag.prefix` flag. For instance, if `-envflag.prefix=VM_`, then env vars must be prepended with `VM_`.
 
-### Configuration with snap package
-
-Snap package for VictoriaMetrics is available [here](https://snapcraft.io/victoriametrics).
-
-Command-line flags for Snap package can be set with following command:
-
-```sh
-echo 'FLAGS="-selfScrapeInterval=10s -search.logSlowQueryDuration=20s"' > $SNAP_DATA/var/snap/victoriametrics/current/extra_flags
-snap restart victoriametrics
-```
-
-Do not change value for `-storageDataPath` flag, because snap package has limited access to host filesystem.
-
-Changing scrape configuration is possible with text editor:
-
-```sh
-vi $SNAP_DATA/var/snap/victoriametrics/current/etc/victoriametrics-scrape-config.yaml
-```
-
-After changes were made, trigger config re-read with the command `curl 127.0.0.1:8428/-/reload`.
-
 ### Running as Windows service
 
 In order to run VictoriaMetrics as a Windows service it is required to create a service configuration for [WinSW](https://github.com/winsw/winsw)
@@ -338,8 +317,8 @@ This allows Grafana to use a more efficient API to get label values.
 Then build graphs and dashboards for the created datasource using [PromQL](https://prometheus.io/docs/prometheus/latest/querying/basics/)
 or [MetricsQL](https://docs.victoriametrics.com/metricsql/).
 
-Alternatively, use VictoriaMetrics [datasource plugin](https://github.com/VictoriaMetrics/grafana-datasource) with support of extra features. 
-See more in [description](https://github.com/VictoriaMetrics/grafana-datasource#victoriametrics-data-source-for-grafana).
+Alternatively, use VictoriaMetrics [datasource plugin](https://github.com/VictoriaMetrics/victoriametrics-datasource) with support of extra features. 
+See more in [description](https://docs.victoriametrics.com/victoriametrics-datasource/).
 
 Creating a datasource may require [specific permissions](https://grafana.com/docs/grafana/latest/administration/data-source-management/).
 If you don't see an option to create a data source - try contacting system administrator.
@@ -702,79 +681,6 @@ The `/api/v1/export` endpoint should return the following response:
 {"metric":{"__name__":"measurement_field2","tag1":"value1","tag2":"value2"},"values":[1.23],"timestamps":[1695902762311]}
 ```
 
-## How to send data from StatsD-compatible clients
-
-VictoriaMetrics supports extended StatsD protocol. Currently, it supports `tags` and `value packing` 
-extensions provided by [dogstatsd](https://docs.datadoghq.com/developers/dogstatsd/datagram_shell). 
-During parsing, metric's `<TYPE>` is added as a special label `__statsd_metric_type__`. 
-
-It is strongly advisable to configure streaming aggregation for each metric type. This process serves two primary 
-objectives: 
-* transformation of the StatsD data model into the VictoriaMetrics data model. VictoriaMetrics requires a consistent 
-interval between data points. 
-* minimizing of the disk space utilization and overall resource consumption during data ingestion.
-
-VictoriaMetrics supports the following metric [types](https://docs.datadoghq.com/metrics/types):
-
-* `c` Counter type.
-* `g`  Gauge type.
-* `ms` Timer type.
-* `m` Meters type.
-* `h` Histogram type.
-* `s` Set type with only numeric values.
-* `d` Distribution type.
-
-_The `Not Assigned` type is not supported due to the ambiguity surrounding its aggregation method. 
-The correct aggregation method cannot be determined for the undefined metric._
-
-Enable Statsd receiver in VictoriaMetrics by setting `-statsdListenAddr` command line flag and configure [stream aggregation](https://docs.victoriametrics.com/stream-aggregation/). 
-For instance, the following command will enable StatsD receiver in VictoriaMetrics on TCP and UDP port `8125`:
-
-```console
-/path/to/victoria-metrics-prod -statsdListenAddr=:8125 -streamAggr.config=statsd_aggr.yaml
-```
-
-Example of stream aggregation config:
-
-```yaml
-# statsd_aggr.yaml
-# `last` output will keep the last sample on `interval`
-# for each series that match `{__statsd_metric_type__="g"}` selector
-- match: '{__statsd_metric_type__="g"}'
-  outputs: [last]
-  interval: 1m
-```
-
-Example for writing data with StatsD plaintext protocol to local VictoriaMetrics using `nc`:
-
-```console
-echo "foo.bar:123|g|#tag1:baz" | nc -N localhost 8125
-```
-
-_An arbitrary number of lines delimited by `\n` (aka newline char) can be sent in one go._
-
-Explicit setting of timestamps is not supported for StatsD protocol. Timestamp is set to the current time when 
-VictoriaMetrics or vmagent receives it.
-
-Once ingested, the data can be read via [/api/v1/export](#how-to-export-data-in-json-line-format) endpoint:
-
-```console
-curl -G 'http://localhost:8428/api/v1/export' -d 'match={__name__=~"foo.*"}'
-```
-
-_Please note, with stream aggregation enabled data will become available only after specified aggregation interval._
-
-The `/api/v1/export` endpoint should return the following response:
-
-```json
-{"metric":{"__name__":"foo.bar:1m_last","__statsd_metric_type__":"g","tag1":"baz"},"values":[123],"timestamps":[1715843939000]}
-```
-
-Some examples of compatible statsd clients:
-- [statsd-instrument](https://github.com/Shopify/statsd-instrument)
-- [dogstatsd-ruby](https://github.com/DataDog/dogstatsd-ruby)
-- [go-statsd-client](https://github.com/cactus/go-statsd-client)
-
 ## How to send data from Graphite-compatible agents such as [StatsD](https://github.com/etsy/statsd)
 
 Enable Graphite receiver in VictoriaMetrics by setting `-graphiteListenAddr` command line flag. For instance,
@@ -793,6 +699,13 @@ Example for writing data with Graphite plaintext protocol to local VictoriaMetri
 echo "foo.bar.baz;tag1=value1;tag2=value2 123 `date +%s`" | nc -N localhost 2003
 ```
 
+To sanitize ingested metric names and labels according to Prometheus naming convention enable
+`-graphite.sanitizeMetricName` cmd-line flag. When enabled, VictoriaMetrics will apply the following modifications:
+- replace `/`,`@`,`*` with `_`;
+- drop `\`;
+- remove redundant dots, e.g: `metric..name` => `metric.name`;
+- replace characters not matching the expression `^a-zA-Z0-9:._` with `_`.
+
 VictoriaMetrics sets the current time if the timestamp is omitted.
 An arbitrary number of lines delimited by `\n` (aka newline char) can be sent in one go.
 After that the data may be read via [/api/v1/export](#how-to-export-data-in-json-line-format) endpoint:
@@ -1205,7 +1118,7 @@ More details may be found [here](https://github.com/VictoriaMetrics/VictoriaMetr
 ## Setting up service
 
 Read [instructions](https://github.com/VictoriaMetrics/VictoriaMetrics/issues/43) on how to set up VictoriaMetrics
-as a service for your OS. A [snap package](https://snapcraft.io/victoriametrics) is available for Ubuntu.
+as a service for your OS.
 
 ## How to work with snapshots
 
@@ -1430,7 +1343,6 @@ Additionally, VictoriaMetrics can accept metrics via the following popular data
 * DataDog `submit metrics` API. See [these docs](#how-to-send-data-from-datadog-agent) for details.
 * InfluxDB line protocol. See [these docs](#how-to-send-data-from-influxdb-compatible-agents-such-as-telegraf) for details.
 * Graphite plaintext protocol. See [these docs](#how-to-send-data-from-graphite-compatible-agents-such-as-statsd) for details.
-* Statsd plaintext protocol. See [these docs](#how-to-send-data-from-statsd-compatible-clients) for details.
 * OpenTelemetry http API. See [these docs](#sending-data-via-opentelemetry) for details.
 * OpenTSDB telnet put protocol. See [these docs](#sending-data-via-telnet-put-protocol) for details.
 * OpenTSDB http `/api/put` protocol. See [these docs](#sending-opentsdb-data-via-http-apiput-requests) for details.
@@ -1624,6 +1536,25 @@ Set HTTP request header `Content-Encoding: gzip` when sending gzip-compressed da
 VictoriaMetrics stores the ingested OpenTelemetry [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples) as is without any transformations.
 Pass `-opentelemetry.usePrometheusNaming` command-line flag to VictoriaMetrics for automatic conversion of metric names and labels into Prometheus-compatible format.
 
+Using the following exporter configuration in the opentelemetry collector will allow you to send metrics into VictoriaMetrics:
+
+```yaml
+exporters:
+  otlphttp/victoriametrics:
+    compression: gzip
+    encoding: proto
+    endpoint: http://<collector/vmagent>.<namespace>.svc.cluster.local:<port>/opentelemetry
+```
+Remember to add the exporter to the desired service pipeline in order to activate the exporter.
+```yaml
+service:
+  pipelines:
+    metrics:
+      exporters:
+        - otlphttp/victoriametrics
+      receivers:
+        - otlp
+```
 See [How to use OpenTelemetry metrics with VictoriaMetrics](https://docs.victoriametrics.com/guides/getting-started-with-opentelemetry/).
 
 ## JSON line format
@@ -2430,7 +2361,8 @@ and [cardinality explorer docs](#cardinality-explorer).
 * If VictoriaMetrics doesn't work because of certain parts are corrupted due to disk errors,
   then just remove directories with broken parts. It is safe removing subdirectories under `<-storageDataPath>/data/{big,small}/YYYY_MM` directories
   when VictoriaMetrics isn't running. This recovers VictoriaMetrics at the cost of data loss stored in the deleted broken parts.
-  In the future, `vmrecover` tool will be created for automatic recovering from such errors.
+  The names of broken parts should be present in the error message. If you see that error message is truncated and doesn't contain all the information
+  try increasing `-loggerMaxArgLen` cmd-line flag to higher values to avoid error messages truncation.
 
 * If you see gaps on the graphs, try resetting the cache by sending request to `/internal/resetRollupResultCache`.
   If this removes gaps on the graphs, then it is likely data with timestamps older than `-search.cacheTimestampOffset`
@@ -2706,10 +2638,39 @@ If you like VictoriaMetrics and want to contribute, then please [read these docs
 
 Report bugs and propose new features [here](https://github.com/VictoriaMetrics/VictoriaMetrics/issues).
 
-## Images in documentation
+## Documentation
+
+VictoriaMetrics documentation is available at [https://docs.victoriametrics.com/](https://docs.victoriametrics.com/).
+It is built from `*.md` files located in [docs](https://github.com/VictoriaMetrics/VictoriaMetrics/tree/master/docs) folder
+and gets automatically updated once changes are merged to [master](https://github.com/VictoriaMetrics/VictoriaMetrics/tree/master) branch.
+To update the documentation follow the steps below:
+- [Fork](https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/about-forks) 
+VictoriaMetrics repo and apply changes to the docs:
+  - To update [the main page](https://docs.victoriametrics.com/) modify [this file](https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/README.md).
+  - To update other pages, apply changes to the corresponding file in [docs folder](https://github.com/VictoriaMetrics/VictoriaMetrics/tree/master/docs).
+- If your changes contain an image then see [images in documentation](https://docs.victoriametrics.com/#images-in-documentation).
+- Once changes are made, execute the command below to finalize and sync the changes:
+```sh
+make docs-sync
+```
+- Create [a pull request](https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/proposing-changes-to-your-work-with-pull-requests/creating-a-pull-request)
+with proposed changes and wait for it to be merged.
+
+Requirements for changes to docs:
+- Keep backward compatibility of existing links. Avoid changing anchors or deleting pages as they could have been
+used or posted in other docs, GitHub issues, stackoverlow answers, etc. 
+- Keep docs simple. Try using as simple wording as possible.
+- Keep docs consistent. When modifying existing docs, verify that other places referencing to this doc are still relevant.
+- Prefer improving the existing docs instead of adding new ones.
+- Use absolute links.
+
+### Images in documentation
 
 Please, keep image size and number of images per single page low. Keep the docs page as lightweight as possible.
 
+Image files must be placed in the same folder as the doc itself and they must have the same prefix as the doc filename.
+For example, all the images for `docs/foo/bar.md` should have filenames starting from `docs/foo/bar`.
+
 If the page needs to have many images, consider using WEB-optimized image format [webp](https://developers.google.com/speed/webp).
 When adding a new doc with many images use `webp` format right away. Or use a Makefile command below to
 convert already existing images at `docs` folder automatically to `web` format:
@@ -2762,7 +2723,7 @@ Pass `-help` to VictoriaMetrics in order to see the list of supported command-li
   -cacheExpireDuration duration
      Items are removed from in-memory caches after they aren't accessed for this duration. Lower values may reduce memory usage at the cost of higher CPU usage. See also -prevCacheRemovalPercent (default 30m0s)
   -configAuthKey value
-     Authorization key for accessing /config page. It must be passed via authKey query arg. It overrides httpAuth.* settings.
+     Authorization key for accessing /config page. It must be passed via authKey query arg. It overrides -httpAuth.*
      Flag value can be read from the given file when using -configAuthKey=file:///abs/path/to/file or -configAuthKey=file://./relative/path/to/file . Flag value can be read from the given http/https url when using -configAuthKey=http://host/path or -configAuthKey=https://host/path
   -csvTrimTimestamp duration
      Trim timestamps when importing csv data to this duration. Minimum practical duration is 1ms. Higher duration (i.e. 1s) may be used for reducing disk space usage for timestamp data (default 1ms)
@@ -2799,7 +2760,7 @@ Pass `-help` to VictoriaMetrics in order to see the list of supported command-li
   -finalMergeDelay duration
      Deprecated: this flag does nothing
   -flagsAuthKey value
-     Auth key for /flags endpoint. It must be passed via authKey query arg. It overrides httpAuth.* settings
+     Auth key for /flags endpoint. It must be passed via authKey query arg. It overrides -httpAuth.*
      Flag value can be read from the given file when using -flagsAuthKey=file:///abs/path/to/file or -flagsAuthKey=file://./relative/path/to/file . Flag value can be read from the given http/https url when using -flagsAuthKey=http://host/path or -flagsAuthKey=https://host/path
   -forceFlushAuthKey value
      authKey, which must be passed in query string to /internal/force_flush pages
@@ -2809,6 +2770,8 @@ Pass `-help` to VictoriaMetrics in order to see the list of supported command-li
      Flag value can be read from the given file when using -forceMergeAuthKey=file:///abs/path/to/file or -forceMergeAuthKey=file://./relative/path/to/file . Flag value can be read from the given http/https url when using -forceMergeAuthKey=http://host/path or -forceMergeAuthKey=https://host/path
   -fs.disableMmap
      Whether to use pread() instead of mmap() for reading data files. By default, mmap() is used for 64-bit arches and pread() is used for 32-bit arches, since they cannot read data files bigger than 2^32 bytes in memory. mmap() is usually faster for reading small data chunks than pread()
+  -graphite.sanitizeMetricName
+     Sanitize metric names for the ingested Graphite data. See https://docs.victoriametrics.com/#how-to-send-data-from-graphite-compatible-agents-such-as-statsd
   -graphiteListenAddr string
      TCP and UDP address to listen for Graphite plaintext data. Usually :2003 must be set. Doesn't work if empty. See also -graphiteListenAddr.useProxyProtocol
   -graphiteListenAddr.useProxyProtocol
@@ -2912,7 +2875,7 @@ Pass `-help` to VictoriaMetrics in order to see the list of supported command-li
      The maximum size in bytes of a single Prometheus remote_write API request
      Supports the following optional suffixes for size values: KB, MB, GB, TB, KiB, MiB, GiB, TiB (default 33554432)
   -maxLabelValueLen int
-     The maximum length of label values in the accepted time series. Longer label values are truncated. In this case the vm_too_long_label_values_total metric at /metrics page is incremented (default 1024)
+     The maximum length of label values in the accepted time series. Longer label values are truncated. In this case the vm_too_long_label_values_total metric at /metrics page is incremented (default 4096)
   -maxLabelsPerTimeseries int
      The maximum number of labels accepted per time series. Superfluous labels are dropped. In this case the vm_metrics_with_dropped_labels_total metric at /metrics page is incremented (default 30)
   -memory.allowedBytes size
@@ -2923,7 +2886,7 @@ Pass `-help` to VictoriaMetrics in order to see the list of supported command-li
   -metrics.exposeMetadata
      Whether to expose TYPE and HELP metadata at the /metrics page, which is exposed at -httpListenAddr . The metadata may be needed when the /metrics page is consumed by systems, which require this information. For example, Managed Prometheus in Google Cloud - https://cloud.google.com/stackdriver/docs/managed-prometheus/troubleshooting#missing-metric-type
   -metricsAuthKey value
-     Auth key for /metrics endpoint. It must be passed via authKey query arg. It overrides httpAuth.* settings
+     Auth key for /metrics endpoint. It must be passed via authKey query arg. It overrides -httpAuth.*
      Flag value can be read from the given file when using -metricsAuthKey=file:///abs/path/to/file or -metricsAuthKey=file://./relative/path/to/file . Flag value can be read from the given http/https url when using -metricsAuthKey=http://host/path or -metricsAuthKey=https://host/path
   -mtls array
      Whether to require valid client certificate for https requests to the corresponding -httpListenAddr . This flag works only if -tls flag is set. See also -mtlsCAFile . This flag is available only in Enterprise binaries. See https://docs.victoriametrics.com/enterprise/
@@ -2954,7 +2917,7 @@ Pass `-help` to VictoriaMetrics in order to see the list of supported command-li
   -opentsdbhttpTrimTimestamp duration
      Trim timestamps for OpenTSDB HTTP data to this duration. Minimum practical duration is 1ms. Higher duration (i.e. 1s) may be used for reducing disk space usage for timestamp data (default 1ms)
   -pprofAuthKey value
-     Auth key for /debug/pprof/* endpoints. It must be passed via authKey query arg. It overrides httpAuth.* settings
+     Auth key for /debug/pprof/* endpoints. It must be passed via authKey query arg. It overrides -httpAuth.*
      Flag value can be read from the given file when using -pprofAuthKey=file:///abs/path/to/file or -pprofAuthKey=file://./relative/path/to/file . Flag value can be read from the given http/https url when using -pprofAuthKey=http://host/path or -pprofAuthKey=https://host/path
   -precisionBits int
      The number of precision bits to store per each value. Lower precision bits improves data compression at the cost of precision loss (default 64)
@@ -3076,7 +3039,7 @@ Pass `-help` to VictoriaMetrics in order to see the list of supported command-li
   -relabelConfig string
      Optional path to a file with relabeling rules, which are applied to all the ingested metrics. The path can point either to local file or to http url. See https://docs.victoriametrics.com/#relabeling for details. The config is reloaded on SIGHUP signal
   -reloadAuthKey value
-     Auth key for /-/reload http endpoint. It must be passed via authKey query arg. It overrides httpAuth.* settings.
+     Auth key for /-/reload http endpoint. It must be passed via authKey query arg. It overrides -httpAuth.*
      Flag value can be read from the given file when using -reloadAuthKey=file:///abs/path/to/file or -reloadAuthKey=file://./relative/path/to/file . Flag value can be read from the given http/https url when using -reloadAuthKey=http://host/path or -reloadAuthKey=https://host/path
   -retentionFilter array
      Retention filter in the format 'filter:retention'. For example, '{env="dev"}:3d' configures the retention for time series with env="dev" label to 3 days. See https://docs.victoriametrics.com/#retention-filters for details. This flag is available only in VictoriaMetrics enterprise. See https://docs.victoriametrics.com/enterprise/
@@ -3093,6 +3056,8 @@ Pass `-help` to VictoriaMetrics in order to see the list of supported command-li
      Whether to disable automatic response cache reset if a sample with timestamp outside -search.cacheTimestampOffset is inserted into VictoriaMetrics
   -search.disableCache
      Whether to disable response caching. This may be useful when ingesting historical data. See https://docs.victoriametrics.com/#backfilling . See also -search.resetRollupResultCacheOnStartup
+  -search.disableImplicitConversion
+     Whether to return an error for queries that rely on implicit subquery conversions, see https://docs.victoriametrics.com/metricsql/#subqueries for details. See also -search.logImplicitConversion
   -search.graphiteMaxPointsPerSeries int
      The maximum number of points per series Graphite render API can return (default 1000000)
   -search.graphiteStorageStep duration
@@ -3101,6 +3066,8 @@ Pass `-help` to VictoriaMetrics in order to see the list of supported command-li
      Whether to ignore match[], extra_filters[] and extra_label query args at /api/v1/labels and /api/v1/label/.../values . This may be useful for decreasing load on VictoriaMetrics when extra filters match too many time series. The downside is that superfluous labels or series could be returned, which do not match the extra filters. See also -search.maxLabelsAPISeries and -search.maxLabelsAPIDuration
   -search.latencyOffset duration
      The time when data points become visible in query results after the collection. It can be overridden on per-query basis via latency_offset arg. Too small value can result in incomplete last points for query results (default 30s)
+  -search.logImplicitConversion
+     Whether to log queries with implicit subquery conversions, see https://docs.victoriametrics.com/metricsql/#subqueries for details. Such conversion can be disabled using -search.disableImplicitConversion
   -search.logQueryMemoryUsage size
      Log query and increment vm_memory_intensive_queries_total metric each time the query requires more memory than specified by this flag. This may help detecting and optimizing heavy queries. Query logging is disabled by default. See also -search.logSlowQueryDuration and -search.maxMemoryPerQuery
      Supports the following optional suffixes for size values: KB, MB, GB, TB, KiB, MiB, GiB, TiB (default 0)
@@ -3123,7 +3090,7 @@ Pass `-help` to VictoriaMetrics in order to see the list of supported command-li
   -search.maxLabelsAPIDuration duration
      The maximum duration for /api/v1/labels, /api/v1/label/.../values and /api/v1/series requests. See also -search.maxLabelsAPISeries and -search.ignoreExtraFiltersAtLabelsAPI (default 5s)
   -search.maxLabelsAPISeries int
-     The maximum number of time series, which could be scanned when searching for the the matching time series at /api/v1/labels and /api/v1/label/.../values. This option allows limiting memory usage and CPU usage. See also -search.maxLabelsAPIDuration, -search.maxTagKeys, -search.maxTagValues and -search.ignoreExtraFiltersAtLabelsAPI (default 1000000)
+     The maximum number of time series, which could be scanned when searching for the matching time series at /api/v1/labels and /api/v1/label/.../values. This option allows limiting memory usage and CPU usage. See also -search.maxLabelsAPIDuration, -search.maxTagKeys, -search.maxTagValues and -search.ignoreExtraFiltersAtLabelsAPI (default 1000000)
   -search.maxLookback duration
      Synonym to -search.lookback-delta from Prometheus. The value is dynamically detected from interval between time series datapoints if not set. It can be overridden on per-query basis via max_lookback arg. See also '-search.maxStalenessInterval' flag, which has the same meaning due to historical reasons
   -search.maxMemoryPerQuery size
@@ -3134,7 +3101,7 @@ Pass `-help` to VictoriaMetrics in order to see the list of supported command-li
   -search.maxPointsSubqueryPerTimeseries int
      The maximum number of points per series, which can be generated by subquery. See https://valyala.medium.com/prometheus-subqueries-in-victoriametrics-9b1492b720b3 (default 100000)
   -search.maxQueryDuration duration
-     The maximum duration for query execution (default 30s)
+     The maximum duration for query execution. It can be overridden on a per-query basis via 'timeout' query arg (default 30s)
   -search.maxQueryLen size
      The maximum search query length in bytes
      Supports the following optional suffixes for size values: KB, MB, GB, TB, KiB, MiB, GiB, TiB (default 16384)
@@ -3206,12 +3173,6 @@ Pass `-help` to VictoriaMetrics in order to see the list of supported command-li
      The following optional suffixes are supported: s (second), m (minute), h (hour), d (day), w (week), y (year). If suffix isn't set, then the duration is counted in months (default 0)
   -sortLabels
      Whether to sort labels for incoming samples before writing them to storage. This may be needed for reducing memory usage at storage when the order of labels in incoming samples is random. For example, if m{k1="v1",k2="v2"} may be sent as m{k2="v2",k1="v1"}. Enabled sorting for labels can slow down ingestion performance a bit
-  -statsd.disableAggregationEnforcement
-    	Whether to disable streaming aggregation requirement check. It's recommended to run statsdServer with pre-configured streaming aggregation to decrease load at database.
-  -statsdListenAddr string
-    	TCP and UDP address to listen for Statsd plaintext data. Usually :8125 must be set. Doesn't work if empty. See also -statsdListenAddr.useProxyProtocol
-  -statsdListenAddr.useProxyProtocol
-    	Whether to use proxy protocol for connections accepted at -statsdListenAddr . See https://www.haproxy.org/download/1.8/doc/proxy-protocol.txt
   -storage.cacheSizeIndexDBDataBlocks size
      Overrides max size for indexdb/dataBlocks cache. See https://docs.victoriametrics.com/single-server-victoriametrics/#cache-tuning
      Supports the following optional suffixes for size values: KB, MB, GB, TB, KiB, MiB, GiB, TiB (default 0)

app/victoria-logs/Makefile

@@ -81,6 +81,9 @@ victoria-logs-linux-ppc64le:
 victoria-logs-linux-s390x:
 	APP_NAME=victoria-logs CGO_ENABLED=0 GOOS=linux GOARCH=s390x $(MAKE) app-local-goos-goarch
 
+victoria-logs-linux-loong64:
+	APP_NAME=victoria-logs CGO_ENABLED=0 GOOS=linux GOARCH=loong64 $(MAKE) app-local-goos-goarch
+
 victoria-logs-linux-386:
 	APP_NAME=victoria-logs CGO_ENABLED=0 GOOS=linux GOARCH=386 $(MAKE) app-local-goos-goarch
 

app/victoria-logs/multiarch/Dockerfile

@@ -1,7 +1,7 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
 ARG certs_image
 ARG root_image
-FROM $certs_image as certs
+FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 
 FROM $root_image

app/victoria-metrics/main_test.go

@@ -3,6 +3,7 @@ package main
 import (
 	"bytes"
 	"encoding/json"
+	"errors"
 	"flag"
 	"fmt"
 	"io"
@@ -117,7 +118,7 @@ func (q *Query) metrics() ([]Metric, error) {
 type QueryInstant struct {
 	Result []struct {
 		Labels map[string]string `json:"metric"`
-		TV     [2]interface{}    `json:"value"`
+		TV     [2]any            `json:"value"`
 	} `json:"result"`
 }
 
@@ -140,7 +141,7 @@ func (q QueryInstant) metrics() ([]Metric, error) {
 type QueryRange struct {
 	Result []struct {
 		Metric map[string]string `json:"metric"`
-		Values [][]interface{}   `json:"values"`
+		Values [][]any           `json:"values"`
 	} `json:"result"`
 }
 
@@ -249,24 +250,24 @@ func TestWriteRead(t *testing.T) {
 
 func testWrite(t *testing.T) {
 	t.Run("prometheus", func(t *testing.T) {
-		for _, test := range readIn("prometheus", t, insertionTime) {
+		for _, test := range readIn("prometheus", insertionTime) {
 			if test.Data == nil {
 				continue
 			}
-			s := newSuite(t)
 			r := testutil.WriteRequest{}
-			s.noError(json.Unmarshal([]byte(strings.Join(test.Data, "\n")), &r.Timeseries))
-			data, err := testutil.Compress(r)
-			s.greaterThan(len(r.Timeseries), 0)
-			if err != nil {
-				t.Errorf("error compressing %v %s", r, err)
-				t.Fail()
+			testData := strings.Join(test.Data, "\n")
+			if err := json.Unmarshal([]byte(testData), &r.Timeseries); err != nil {
+				panic(fmt.Errorf("BUG: cannot unmarshal TimeSeries: %s\ntest data\n%s", err, testData))
 			}
+			if n := len(r.Timeseries); n <= 0 {
+				panic(fmt.Errorf("BUG: expecting non-empty Timeseries in test data:\n%s", testData))
+			}
+			data := testutil.Compress(r)
 			httpWrite(t, testPromWriteHTTPPath, test.InsertQuery, bytes.NewBuffer(data))
 		}
 	})
 	t.Run("csv", func(t *testing.T) {
-		for _, test := range readIn("csv", t, insertionTime) {
+		for _, test := range readIn("csv", insertionTime) {
 			if test.Data == nil {
 				continue
 			}
@@ -275,7 +276,7 @@ func testWrite(t *testing.T) {
 	})
 
 	t.Run("influxdb", func(t *testing.T) {
-		for _, x := range readIn("influxdb", t, insertionTime) {
+		for _, x := range readIn("influxdb", insertionTime) {
 			test := x
 			t.Run(test.Name, func(t *testing.T) {
 				t.Parallel()
@@ -284,7 +285,7 @@ func testWrite(t *testing.T) {
 		}
 	})
 	t.Run("graphite", func(t *testing.T) {
-		for _, x := range readIn("graphite", t, insertionTime) {
+		for _, x := range readIn("graphite", insertionTime) {
 			test := x
 			t.Run(test.Name, func(t *testing.T) {
 				t.Parallel()
@@ -293,7 +294,7 @@ func testWrite(t *testing.T) {
 		}
 	})
 	t.Run("opentsdb", func(t *testing.T) {
-		for _, x := range readIn("opentsdb", t, insertionTime) {
+		for _, x := range readIn("opentsdb", insertionTime) {
 			test := x
 			t.Run(test.Name, func(t *testing.T) {
 				t.Parallel()
@@ -302,7 +303,7 @@ func testWrite(t *testing.T) {
 		}
 	})
 	t.Run("opentsdbhttp", func(t *testing.T) {
-		for _, x := range readIn("opentsdbhttp", t, insertionTime) {
+		for _, x := range readIn("opentsdbhttp", insertionTime) {
 			test := x
 			t.Run(test.Name, func(t *testing.T) {
 				t.Parallel()
@@ -316,7 +317,7 @@ func testWrite(t *testing.T) {
 func testRead(t *testing.T) {
 	for _, engine := range []string{"csv", "prometheus", "graphite", "opentsdb", "influxdb", "opentsdbhttp"} {
 		t.Run(engine, func(t *testing.T) {
-			for _, x := range readIn(engine, t, insertionTime) {
+			for _, x := range readIn(engine, insertionTime) {
 				test := x
 				t.Run(test.Name, func(t *testing.T) {
 					t.Parallel()
@@ -365,11 +366,10 @@ func testRead(t *testing.T) {
 	}
 }
 
-func readIn(readFor string, t *testing.T, insertTime time.Time) []test {
-	t.Helper()
-	s := newSuite(t)
+func readIn(readFor string, insertTime time.Time) []test {
+	testDir := filepath.Join(testFixturesDir, readFor)
 	var tt []test
-	s.noError(filepath.Walk(filepath.Join(testFixturesDir, readFor), func(path string, _ os.FileInfo, err error) error {
+	err := filepath.Walk(testDir, func(path string, _ os.FileInfo, err error) error {
 		if err != nil {
 			return err
 		}
@@ -377,84 +377,129 @@ func readIn(readFor string, t *testing.T, insertTime time.Time) []test {
 			return nil
 		}
 		b, err := os.ReadFile(path)
-		s.noError(err)
+		if err != nil {
+			panic(fmt.Errorf("BUG: cannot read %s: %s", path, err))
+		}
 		item := test{}
-		s.noError(json.Unmarshal(b, &item))
+		if err := json.Unmarshal(b, &item); err != nil {
+			panic(fmt.Errorf("cannot parse %T from %s: %s; data:\n%s", &item, path, err, b))
+		}
 		for i := range item.Data {
 			item.Data[i] = testutil.PopulateTimeTplString(item.Data[i], insertTime)
 		}
 		tt = append(tt, item)
 		return nil
-	}))
+	})
+	if err != nil {
+		panic(fmt.Errorf("BUG: cannot read test data at %s: %w", testDir, err))
+	}
 	if len(tt) == 0 {
-		t.Fatalf("no test found in %s", filepath.Join(testFixturesDir, readFor))
+		panic(fmt.Errorf("BUG: no tests found in %s", testDir))
 	}
 	return tt
 }
 
 func httpWrite(t *testing.T, address, query string, r io.Reader) {
 	t.Helper()
-	s := newSuite(t)
-	resp, err := http.Post(address+query, "", r)
-	s.noError(err)
-	s.noError(resp.Body.Close())
-	s.equalInt(resp.StatusCode, 204)
+
+	requestURL := address + query
+	resp, err := http.Post(requestURL, "", r)
+	if err != nil {
+		t.Fatalf("cannot send request to %s: %s", requestURL, err)
+	}
+	_ = resp.Body.Close()
+	if resp.StatusCode != 204 {
+		t.Fatalf("unexpected status code received from %s; got %d; want 204", requestURL, resp.StatusCode)
+	}
 }
 
-func tcpWrite(t *testing.T, address string, data string) {
+func tcpWrite(t *testing.T, address, data string) {
 	t.Helper()
-	s := newSuite(t)
+
 	conn, err := net.Dial("tcp", address)
-	s.noError(err)
+	if err != nil {
+		t.Fatalf("cannot dial %s: %s", address, err)
+	}
 	defer func() {
 		_ = conn.Close()
 	}()
 	n, err := conn.Write([]byte(data))
-	s.noError(err)
-	s.equalInt(n, len(data))
+	if err != nil {
+		t.Fatalf("cannot write %d bytes to %s: %s", len(data), address, err)
+	}
+	if n != len(data) {
+		panic(fmt.Errorf("BUG: conn.Write() returned unexpected number of written bytes to %s; got %d; want %d", address, n, len(data)))
+	}
 }
 
 func httpReadMetrics(t *testing.T, address, query string) []Metric {
 	t.Helper()
-	s := newSuite(t)
-	resp, err := http.Get(address + query)
-	s.noError(err)
+
+	requestURL := address + query
+	resp, err := http.Get(requestURL)
+	if err != nil {
+		t.Fatalf("cannot send request to %s: %s", requestURL, err)
+	}
 	defer func() {
 		_ = resp.Body.Close()
 	}()
-	s.equalInt(resp.StatusCode, 200)
+	if resp.StatusCode != 200 {
+		t.Fatalf("unexpected status code received from %s; got %d; want 200", requestURL, resp.StatusCode)
+	}
+
 	var rows []Metric
-	for dec := json.NewDecoder(resp.Body); dec.More(); {
+	dec := json.NewDecoder(resp.Body)
+	for {
 		var row Metric
-		s.noError(dec.Decode(&row))
+		err := dec.Decode(&row)
+		if err != nil {
+			if errors.Is(err, io.EOF) {
+				return rows
+			}
+			t.Fatalf("cannot decode %T from response received from %s: %s", &row, requestURL, err)
+		}
 		rows = append(rows, row)
 	}
-	return rows
 }
 
-func httpReadStruct(t *testing.T, address, query string, dst interface{}) {
+func httpReadStruct(t *testing.T, address, query string, dst any) {
 	t.Helper()
-	s := newSuite(t)
-	resp, err := http.Get(address + query)
-	s.noError(err)
+
+	requestURL := address + query
+	resp, err := http.Get(requestURL)
+	if err != nil {
+		t.Fatalf("cannot send request to %s: %s", requestURL, err)
+	}
 	defer func() {
 		_ = resp.Body.Close()
 	}()
-	s.equalInt(resp.StatusCode, 200)
-	s.noError(json.NewDecoder(resp.Body).Decode(dst))
+	if resp.StatusCode != 200 {
+		t.Fatalf("unexpected status code received from %s; got %d; want 200", requestURL, resp.StatusCode)
+	}
+	err = json.NewDecoder(resp.Body).Decode(dst)
+	if err != nil {
+		t.Fatalf("cannot decode %T from response received from %s: %s", dst, requestURL, err)
+	}
 }
 
 func httpReadData(t *testing.T, address, query string) []byte {
 	t.Helper()
-	s := newSuite(t)
-	resp, err := http.Get(address + query)
-	s.noError(err)
+
+	requestURL := address + query
+	resp, err := http.Get(requestURL)
+	if err != nil {
+		t.Fatalf("cannot send request to %s: %s", requestURL, err)
+	}
 	defer func() {
 		_ = resp.Body.Close()
 	}()
-	s.equalInt(resp.StatusCode, 200)
+	if resp.StatusCode != 200 {
+		t.Fatalf("unexpected status code received from %s; got %d; want 200", requestURL, resp.StatusCode)
+	}
 	data, err := io.ReadAll(resp.Body)
-	s.noError(err)
+	if err != nil {
+		t.Fatalf("cannot read response from %s: %s", requestURL, err)
+	}
 	return data
 }
 
@@ -503,34 +548,6 @@ func removeIfFoundSeries(r map[string]string, contains []map[string]string) []ma
 	return contains
 }
 
-type suite struct{ t *testing.T }
-
-func newSuite(t *testing.T) *suite { return &suite{t: t} }
-
-func (s *suite) noError(err error) {
-	s.t.Helper()
-	if err != nil {
-		s.t.Errorf("unexpected error %v", err)
-		s.t.FailNow()
-	}
-}
-
-func (s *suite) equalInt(a, b int) {
-	s.t.Helper()
-	if a != b {
-		s.t.Errorf("%d not equal %d", a, b)
-		s.t.FailNow()
-	}
-}
-
-func (s *suite) greaterThan(a, b int) {
-	s.t.Helper()
-	if a <= b {
-		s.t.Errorf("%d less or equal then %d", a, b)
-		s.t.FailNow()
-	}
-}
-
 func TestImportJSONLines(t *testing.T) {
 	f := func(labelsCount, labelLen int) {
 		t.Helper()

app/victoria-metrics/multiarch/Dockerfile

@@ -1,7 +1,7 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
 ARG certs_image
 ARG root_image
-FROM $certs_image as certs
+FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 
 FROM $root_image

app/victoria-metrics/test/prom_writer.go

@@ -1,12 +1,16 @@
 package test
 
-import "github.com/golang/snappy"
+import (
+	"fmt"
+
+	"github.com/golang/snappy"
+)
 
 // Compress marshals and compresses wr.
-func Compress(wr WriteRequest) ([]byte, error) {
+func Compress(wr WriteRequest) []byte {
 	data, err := wr.Marshal()
 	if err != nil {
-		return nil, err
+		panic(fmt.Errorf("BUG: cannot compress WriteRequest: %s", err))
 	}
-	return snappy.Encode(nil, data), nil
+	return snappy.Encode(nil, data)
 }

app/vlinsert/insertutils/common_params.go

@@ -2,6 +2,8 @@ package insertutils
 
 import (
 	"net/http"
+	"sync"
+	"time"
 
 	"github.com/VictoriaMetrics/metrics"
 
@@ -10,6 +12,7 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputils"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/timeutil"
 )
 
 // CommonParams contains common HTTP parameters used by log ingestion APIs.
@@ -100,12 +103,46 @@ type LogMessageProcessor interface {
 }
 
 type logMessageProcessor struct {
+	mu            sync.Mutex
+	wg            sync.WaitGroup
+	stopCh        chan struct{}
+	lastFlushTime time.Time
+
 	cp *CommonParams
 	lr *logstorage.LogRows
 }
 
+func (lmp *logMessageProcessor) initPeriodicFlush() {
+	lmp.lastFlushTime = time.Now()
+
+	lmp.wg.Add(1)
+	go func() {
+		defer lmp.wg.Done()
+
+		d := timeutil.AddJitterToDuration(time.Second)
+		ticker := time.NewTicker(d)
+		defer ticker.Stop()
+
+		for {
+			select {
+			case <-lmp.stopCh:
+				return
+			case <-ticker.C:
+				lmp.mu.Lock()
+				if time.Since(lmp.lastFlushTime) >= d {
+					lmp.flushLocked()
+				}
+				lmp.mu.Unlock()
+			}
+		}
+	}()
+}
+
 // AddRow adds new log message to lmp with the given timestamp and fields.
 func (lmp *logMessageProcessor) AddRow(timestamp int64, fields []logstorage.Field) {
+	lmp.mu.Lock()
+	defer lmp.mu.Unlock()
+
 	if len(fields) > *MaxFieldsPerLine {
 		rf := logstorage.RowFormatter(fields)
 		logger.Warnf("dropping log line with %d fields; it exceeds -insert.maxFieldsPerLine=%d; %s", len(fields), *MaxFieldsPerLine, rf)
@@ -122,29 +159,41 @@ func (lmp *logMessageProcessor) AddRow(timestamp int64, fields []logstorage.Fiel
 		return
 	}
 	if lmp.lr.NeedFlush() {
-		lmp.flush()
+		lmp.flushLocked()
 	}
 }
 
-func (lmp *logMessageProcessor) flush() {
+// flushLocked must be called under locked lmp.mu.
+func (lmp *logMessageProcessor) flushLocked() {
+	lmp.lastFlushTime = time.Now()
 	vlstorage.MustAddRows(lmp.lr)
 	lmp.lr.ResetKeepSettings()
 }
 
 // MustClose flushes the remaining data to the underlying storage and closes lmp.
 func (lmp *logMessageProcessor) MustClose() {
-	lmp.flush()
+	close(lmp.stopCh)
+	lmp.wg.Wait()
+
+	lmp.flushLocked()
 	logstorage.PutLogRows(lmp.lr)
 	lmp.lr = nil
 }
 
 // NewLogMessageProcessor returns new LogMessageProcessor for the given cp.
+//
+// MustClose() must be called on the returned LogMessageProcessor when it is no longer needed.
 func (cp *CommonParams) NewLogMessageProcessor() LogMessageProcessor {
 	lr := logstorage.GetLogRows(cp.StreamFields, cp.IgnoreFields)
-	return &logMessageProcessor{
+	lmp := &logMessageProcessor{
 		cp: cp,
 		lr: lr,
+
+		stopCh: make(chan struct{}),
 	}
+	lmp.initPeriodicFlush()
+
+	return lmp
 }
 
 var rowsDroppedTotalDebug = metrics.NewCounter(`vl_rows_dropped_total{reason="debug"}`)

app/vlinsert/insertutils/timestamp.go

@@ -19,12 +19,12 @@ func ExtractTimestampRFC3339NanoFromFields(timeField string, fields []logstorage
 		if f.Name != timeField {
 			continue
 		}
+		if f.Value == "" || f.Value == "0" {
+			return time.Now().UnixNano(), nil
+		}
 		nsecs, ok := logstorage.TryParseTimestampRFC3339Nano(f.Value)
 		if !ok {
-			if f.Value == "0" || f.Value == "" {
-				return time.Now().UnixNano(), nil
-			}
-			return time.Now().UnixNano(), fmt.Errorf("cannot unmarshal iso8601 timestamp from %s=%q", timeField, f.Value)
+			return 0, fmt.Errorf("cannot unmarshal rfc3339 timestamp from %s=%q", timeField, f.Value)
 		}
 		f.Value = ""
 		return nsecs, nil

app/vlinsert/insertutils/timestamp_test.go

@@ -0,0 +1,75 @@
+package insertutils
+
+import (
+	"testing"
+
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
+)
+
+func TestExtractTimestampRFC3339NanoFromFields_Success(t *testing.T) {
+	f := func(timeField string, fields []logstorage.Field, nsecsExpected int64) {
+		t.Helper()
+
+		nsecs, err := ExtractTimestampRFC3339NanoFromFields(timeField, fields)
+		if err != nil {
+			t.Fatalf("unexpected error: %s", err)
+		}
+		if nsecs != nsecsExpected {
+			t.Fatalf("unexpected nsecs; got %d; want %d", nsecs, nsecsExpected)
+		}
+
+		for _, f := range fields {
+			if f.Name == timeField {
+				if f.Value != "" {
+					t.Fatalf("unexpected value for field %s; got %q; want %q", timeField, f.Value, "")
+				}
+			}
+		}
+	}
+
+	f("time", []logstorage.Field{
+		{Name: "foo", Value: "bar"},
+		{Name: "time", Value: "2024-06-18T23:37:20Z"},
+	}, 1718753840000000000)
+
+	f("time", []logstorage.Field{
+		{Name: "foo", Value: "bar"},
+		{Name: "time", Value: "2024-06-18T23:37:20+08:00"},
+	}, 1718725040000000000)
+
+	f("time", []logstorage.Field{
+		{Name: "foo", Value: "bar"},
+		{Name: "time", Value: "2024-06-18T23:37:20.123-05:30"},
+	}, 1718773640123000000)
+
+	f("time", []logstorage.Field{
+		{Name: "time", Value: "2024-06-18T23:37:20.123456789-05:30"},
+		{Name: "foo", Value: "bar"},
+	}, 1718773640123456789)
+}
+
+func TestExtractTimestampRFC3339NanoFromFields_Error(t *testing.T) {
+	f := func(s string) {
+		t.Helper()
+
+		fields := []logstorage.Field{
+			{Name: "time", Value: s},
+		}
+		nsecs, err := ExtractTimestampRFC3339NanoFromFields("time", fields)
+		if err == nil {
+			t.Fatalf("expecting non-nil error")
+		}
+		if nsecs != 0 {
+			t.Fatalf("unexpected nsecs; got %d; want %d", nsecs, 0)
+		}
+	}
+
+	f("foobar")
+
+	// no Z at the end
+	f("2024-06-18T23:37:20")
+
+	// incomplete time
+	f("2024-06-18")
+	f("2024-06-18T23:37")
+}

app/vlinsert/loki/loki_protobuf.go

@@ -79,12 +79,14 @@ func parseProtobufRequest(data []byte, lmp insertutils.LogMessageProcessor) (int
 	req := getPushRequest()
 	defer putPushRequest(req)
 
-	err = req.Unmarshal(bb.B)
+	err = req.UnmarshalProtobuf(bb.B)
 	if err != nil {
 		return 0, fmt.Errorf("cannot parse request body: %w", err)
 	}
 
-	var commonFields []logstorage.Field
+	fields := getFields()
+	defer putFields(fields)
+
 	rowsIngested := 0
 	streams := req.Streams
 	currentTimestamp := time.Now().UnixNano()
@@ -92,30 +94,60 @@ func parseProtobufRequest(data []byte, lmp insertutils.LogMessageProcessor) (int
 		stream := &streams[i]
 		// st.Labels contains labels for the stream.
 		// Labels are same for all entries in the stream.
-		commonFields, err = parsePromLabels(commonFields[:0], stream.Labels)
+		fields.fields, err = parsePromLabels(fields.fields[:0], stream.Labels)
 		if err != nil {
 			return rowsIngested, fmt.Errorf("cannot parse stream labels %q: %w", stream.Labels, err)
 		}
-		fields := commonFields
+		commonFieldsLen := len(fields.fields)
 
 		entries := stream.Entries
 		for j := range entries {
-			entry := &entries[j]
-			fields = append(fields[:len(commonFields)], logstorage.Field{
+			e := &entries[j]
+			fields.fields = fields.fields[:commonFieldsLen]
+
+			for _, lp := range e.StructuredMetadata {
+				fields.fields = append(fields.fields, logstorage.Field{
+					Name:  lp.Name,
+					Value: lp.Value,
+				})
+			}
+
+			fields.fields = append(fields.fields, logstorage.Field{
 				Name:  "_msg",
-				Value: entry.Line,
+				Value: e.Line,
 			})
-			ts := entry.Timestamp.UnixNano()
+
+			ts := e.Timestamp.UnixNano()
 			if ts == 0 {
 				ts = currentTimestamp
 			}
-			lmp.AddRow(ts, fields)
+
+			lmp.AddRow(ts, fields.fields)
 		}
 		rowsIngested += len(stream.Entries)
 	}
 	return rowsIngested, nil
 }
 
+func getFields() *fields {
+	v := fieldsPool.Get()
+	if v == nil {
+		return &fields{}
+	}
+	return v.(*fields)
+}
+
+func putFields(f *fields) {
+	f.fields = f.fields[:0]
+	fieldsPool.Put(f)
+}
+
+var fieldsPool sync.Pool
+
+type fields struct {
+	fields []logstorage.Field
+}
+
 // parsePromLabels parses log fields in Prometheus text exposition format from s, appends them to dst and returns the result.
 //
 // See test data of promtail for examples: https://github.com/grafana/loki/blob/a24ef7b206e0ca63ee74ca6ecb0a09b745cd2258/pkg/push/types_test.go
@@ -181,6 +213,6 @@ func getPushRequest() *PushRequest {
 }
 
 func putPushRequest(req *PushRequest) {
-	req.Reset()
+	req.reset()
 	pushReqsPool.Put(req)
 }

app/vlinsert/loki/loki_protobuf_test.go

@@ -36,7 +36,7 @@ func (tlp *testLogMessageProcessor) AddRow(timestamp int64, fields []logstorage.
 		Entries: []Entry{
 			{
 				Timestamp: time.Unix(0, timestamp),
-				Line:      msg,
+				Line:      strings.Clone(msg),
 			},
 		},
 	})
@@ -58,10 +58,7 @@ func TestParseProtobufRequest_Success(t *testing.T) {
 			t.Fatalf("unexpected number of streams; got %d; want %d", len(tlp.pr.Streams), n)
 		}
 
-		data, err := tlp.pr.Marshal()
-		if err != nil {
-			t.Fatalf("unexpected error when marshaling PushRequest: %s", err)
-		}
+		data := tlp.pr.MarshalProtobuf(nil)
 		encodedData := snappy.Encode(nil, data)
 
 		tlp2 := &insertutils.TestLogMessageProcessor{}

app/vlinsert/loki/loki_protobuf_timing_test.go

@@ -9,6 +9,7 @@ import (
 	"github.com/golang/snappy"
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
 )
 
 func BenchmarkParseProtobufRequest(b *testing.B) {
@@ -38,29 +39,47 @@ func benchmarkParseProtobufRequest(b *testing.B, streams, rows, labels int) {
 	})
 }
 
-func getProtobufBody(streams, rows, labels int) []byte {
-	var pr PushRequest
-
-	for i := 0; i < streams; i++ {
-		var st Stream
-
-		st.Labels = `{`
-		for j := 0; j < labels; j++ {
-			st.Labels += `label_` + strconv.Itoa(j) + `="value_` + strconv.Itoa(j) + `"`
-			if j < labels-1 {
-				st.Labels += `,`
+func getProtobufBody(streamsCount, rowsCount, labelsCount int) []byte {
+	var b []byte
+	var entries []Entry
+	streams := make([]Stream, streamsCount)
+	for i := range streams {
+		b = b[:0]
+		b = append(b, '{')
+		for j := 0; j < labelsCount; j++ {
+			b = append(b, "label_"...)
+			b = strconv.AppendInt(b, int64(j), 10)
+			b = append(b, `="value_`...)
+			b = strconv.AppendInt(b, int64(j), 10)
+			b = append(b, '"')
+			if j < labelsCount-1 {
+				b = append(b, ',')
 			}
 		}
-		st.Labels += `}`
+		b = append(b, '}')
+		labels := string(b)
 
-		for j := 0; j < rows; j++ {
-			st.Entries = append(st.Entries, Entry{Timestamp: time.Now(), Line: "value_" + strconv.Itoa(j)})
+		var rowsBuf []byte
+		entriesLen := len(entries)
+		for j := 0; j < rowsCount; j++ {
+			rowsBufLen := len(rowsBuf)
+			rowsBuf = append(rowsBuf, "value_"...)
+			rowsBuf = strconv.AppendInt(rowsBuf, int64(j), 10)
+			entries = append(entries, Entry{
+				Timestamp: time.Now(),
+				Line:      bytesutil.ToUnsafeString(rowsBuf[rowsBufLen:]),
+			})
 		}
 
-		pr.Streams = append(pr.Streams, st)
+		st := &streams[i]
+		st.Labels = labels
+		st.Entries = entries[entriesLen:]
+	}
+	pr := PushRequest{
+		Streams: streams,
 	}
 
-	body, _ := pr.Marshal()
+	body := pr.MarshalProtobuf(nil)
 	encodedBody := snappy.Encode(nil, body)
 
 	return encodedBody

app/vlinsert/loki/pb.go

@@ -0,0 +1,302 @@
+// Code generated by protoc-gen-gogo. DO NOT EDIT.
+// source: push_request.proto
+// source: https://raw.githubusercontent.com/grafana/loki/main/pkg/push/push_request.proto
+// Licensed under the Apache License, Version 2.0 (the "License");
+// https://github.com/grafana/loki/blob/main/pkg/push/LICENSE
+
+package loki
+
+import (
+	"fmt"
+	"time"
+
+	"github.com/VictoriaMetrics/easyproto"
+)
+
+var mp easyproto.MarshalerPool
+
+// PushRequest represents Loki PushRequest
+//
+// See https://github.com/grafana/loki/blob/4220737a52da7ab6c9346b12d5a5d7bedbcd641d/pkg/push/push.proto#L14C1-L14C20
+type PushRequest struct {
+	Streams []Stream
+
+	entriesBuf   []Entry
+	labelPairBuf []LabelPair
+}
+
+func (pr *PushRequest) reset() {
+	pr.Streams = pr.Streams[:0]
+
+	pr.entriesBuf = pr.entriesBuf[:0]
+	pr.labelPairBuf = pr.labelPairBuf[:0]
+}
+
+// UnmarshalProtobuf unmarshals pr from protobuf message at src.
+//
+// pr remains valid until src is modified.
+func (pr *PushRequest) UnmarshalProtobuf(src []byte) error {
+	pr.reset()
+	var err error
+	pr.entriesBuf, pr.labelPairBuf, err = pr.unmarshalProtobuf(pr.entriesBuf, pr.labelPairBuf, src)
+	return err
+}
+
+// MarshalProtobuf marshals r to protobuf message, appends it to dst and returns the result.
+func (pr *PushRequest) MarshalProtobuf(dst []byte) []byte {
+	m := mp.Get()
+	pr.marshalProtobuf(m.MessageMarshaler())
+	dst = m.Marshal(dst)
+	mp.Put(m)
+	return dst
+}
+
+func (pr *PushRequest) marshalProtobuf(mm *easyproto.MessageMarshaler) {
+	for _, s := range pr.Streams {
+		s.marshalProtobuf(mm.AppendMessage(1))
+	}
+}
+
+func (pr *PushRequest) unmarshalProtobuf(entriesBuf []Entry, labelPairBuf []LabelPair, src []byte) ([]Entry, []LabelPair, error) {
+	// message PushRequest {
+	//   repeated Stream streams = 1;
+	// }
+	var err error
+	var fc easyproto.FieldContext
+	for len(src) > 0 {
+		src, err = fc.NextField(src)
+		if err != nil {
+			return entriesBuf, labelPairBuf, fmt.Errorf("cannot read next field in PushRequest: %w", err)
+		}
+		switch fc.FieldNum {
+		case 1:
+			data, ok := fc.MessageData()
+			if !ok {
+				return entriesBuf, labelPairBuf, fmt.Errorf("cannot read Stream data")
+			}
+			pr.Streams = append(pr.Streams, Stream{})
+			s := &pr.Streams[len(pr.Streams)-1]
+			entriesBuf, labelPairBuf, err = s.unmarshalProtobuf(entriesBuf, labelPairBuf, data)
+			if err != nil {
+				return entriesBuf, labelPairBuf, fmt.Errorf("cannot unmarshal Stream: %w", err)
+			}
+		}
+	}
+	return entriesBuf, labelPairBuf, nil
+}
+
+// Stream represents Loki stream.
+//
+// See https://github.com/grafana/loki/blob/4220737a52da7ab6c9346b12d5a5d7bedbcd641d/pkg/push/push.proto#L23
+type Stream struct {
+	Labels  string
+	Entries []Entry
+}
+
+func (s *Stream) marshalProtobuf(mm *easyproto.MessageMarshaler) {
+	mm.AppendString(1, s.Labels)
+	for _, e := range s.Entries {
+		e.marshalProtobuf(mm.AppendMessage(2))
+	}
+}
+
+func (s *Stream) unmarshalProtobuf(entriesBuf []Entry, labelPairBuf []LabelPair, src []byte) ([]Entry, []LabelPair, error) {
+	// message Stream {
+	//   string labels = 1;
+	//   repeated Entry entries = 2;
+	// }
+	var err error
+	var fc easyproto.FieldContext
+	entriesBufLen := len(entriesBuf)
+	for len(src) > 0 {
+		src, err = fc.NextField(src)
+		if err != nil {
+			return entriesBuf, labelPairBuf, fmt.Errorf("cannot read next field in Stream: %w", err)
+		}
+		switch fc.FieldNum {
+		case 1:
+			labels, ok := fc.String()
+			if !ok {
+				return entriesBuf, labelPairBuf, fmt.Errorf("cannot read labels")
+			}
+			s.Labels = labels
+		case 2:
+			data, ok := fc.MessageData()
+			if !ok {
+				return entriesBuf, labelPairBuf, fmt.Errorf("cannot read Entry data")
+			}
+			entriesBuf = append(entriesBuf, Entry{})
+			e := &entriesBuf[len(entriesBuf)-1]
+			labelPairBuf, err = e.unmarshalProtobuf(labelPairBuf, data)
+			if err != nil {
+				return entriesBuf, labelPairBuf, fmt.Errorf("cannot unmarshal Entry: %w", err)
+			}
+		}
+	}
+	s.Entries = entriesBuf[entriesBufLen:]
+	return entriesBuf, labelPairBuf, nil
+}
+
+// Entry represents Loki entry.
+//
+// See https://github.com/grafana/loki/blob/4220737a52da7ab6c9346b12d5a5d7bedbcd641d/pkg/push/push.proto#L38
+type Entry struct {
+	Timestamp          time.Time
+	Line               string
+	StructuredMetadata []LabelPair
+}
+
+func (e *Entry) marshalProtobuf(mm *easyproto.MessageMarshaler) {
+	marshalTime(mm, 1, e.Timestamp)
+	mm.AppendString(2, e.Line)
+	for _, lp := range e.StructuredMetadata {
+		lp.marshalProtobuf(mm.AppendMessage(3))
+	}
+}
+
+func (e *Entry) unmarshalProtobuf(labelPairBuf []LabelPair, src []byte) ([]LabelPair, error) {
+	// message Entry {
+	//   Timestamp timestamp = 1;
+	//   string line = 2;
+	//   repeated LabelPair structuredMetadata = 3;
+	// }
+	var err error
+	var fc easyproto.FieldContext
+	labelPairBufLen := len(labelPairBuf)
+	for len(src) > 0 {
+		src, err = fc.NextField(src)
+		if err != nil {
+			return labelPairBuf, fmt.Errorf("cannot read next field in Entry: %w", err)
+		}
+		switch fc.FieldNum {
+		case 1:
+			data, ok := fc.MessageData()
+			if !ok {
+				return labelPairBuf, fmt.Errorf("cannot read Timestamp data")
+			}
+			timestamp, err := unmarshalTime(data)
+			if err != nil {
+				return labelPairBuf, fmt.Errorf("cannot unmarshal Timestamp: %w", err)
+			}
+			e.Timestamp = timestamp
+		case 2:
+			line, ok := fc.String()
+			if !ok {
+				return labelPairBuf, fmt.Errorf("cannot read Line")
+			}
+			e.Line = line
+		case 3:
+			data, ok := fc.MessageData()
+			if !ok {
+				return labelPairBuf, fmt.Errorf("cannot read StructuredMetadata")
+			}
+			labelPairBuf = append(labelPairBuf, LabelPair{})
+			lp := &labelPairBuf[len(labelPairBuf)-1]
+			if err := lp.unmarshalProtobuf(data); err != nil {
+				return labelPairBuf, fmt.Errorf("cannot unmarshal StructuredMetadata: %w", err)
+			}
+		}
+	}
+	e.StructuredMetadata = labelPairBuf[labelPairBufLen:]
+	return labelPairBuf, nil
+}
+
+// LabelPair represents Loki label pair.
+//
+// See https://github.com/grafana/loki/blob/4220737a52da7ab6c9346b12d5a5d7bedbcd641d/pkg/push/push.proto#L33
+type LabelPair struct {
+	Name  string
+	Value string
+}
+
+func (lp *LabelPair) marshalProtobuf(mm *easyproto.MessageMarshaler) {
+	mm.AppendString(1, lp.Name)
+	mm.AppendString(2, lp.Value)
+}
+
+func (lp *LabelPair) unmarshalProtobuf(src []byte) (err error) {
+	// message LabelPair {
+	//   string name = 1;
+	//   string value = 2;
+	// }
+	var fc easyproto.FieldContext
+	for len(src) > 0 {
+		src, err = fc.NextField(src)
+		if err != nil {
+			return fmt.Errorf("cannot read next field in LabelPair: %w", err)
+		}
+		switch fc.FieldNum {
+		case 1:
+			name, ok := fc.String()
+			if !ok {
+				return fmt.Errorf("cannot read name")
+			}
+			lp.Name = name
+		case 2:
+			value, ok := fc.String()
+			if !ok {
+				return fmt.Errorf("cannot unmarshal value")
+			}
+			lp.Value = value
+		}
+	}
+	return nil
+}
+
+func marshalTime(mm *easyproto.MessageMarshaler, fieldNum uint32, timestamp time.Time) {
+	nsecs := timestamp.UnixNano()
+	ts := Timestamp{
+		Seconds: nsecs / 1e9,
+		Nanos:   int32(nsecs % 1e9),
+	}
+	ts.marshalProtobuf(mm.AppendMessage(fieldNum))
+}
+
+func unmarshalTime(src []byte) (time.Time, error) {
+	var ts Timestamp
+	if err := ts.unmarshalProtobuf(src); err != nil {
+		return time.Time{}, err
+	}
+	timestamp := time.Unix(ts.Seconds, int64(ts.Nanos)).UTC()
+	return timestamp, nil
+}
+
+// Timestamp is protobuf well-known timestamp type.
+type Timestamp struct {
+	Seconds int64
+	Nanos   int32
+}
+
+func (ts *Timestamp) marshalProtobuf(mm *easyproto.MessageMarshaler) {
+	mm.AppendInt64(1, ts.Seconds)
+	mm.AppendInt32(2, ts.Nanos)
+}
+
+func (ts *Timestamp) unmarshalProtobuf(src []byte) (err error) {
+	// message Timestamp {
+	//   int64 seconds = 1;
+	//   int32 nanos = 2;
+	// }
+	var fc easyproto.FieldContext
+	for len(src) > 0 {
+		src, err = fc.NextField(src)
+		if err != nil {
+			return fmt.Errorf("cannot read next field in Timestamp: %w", err)
+		}
+		switch fc.FieldNum {
+		case 1:
+			seconds, ok := fc.Int64()
+			if !ok {
+				return fmt.Errorf("cannot read Seconds")
+			}
+			ts.Seconds = seconds
+		case 2:
+			nanos, ok := fc.Int32()
+			if !ok {
+				return fmt.Errorf("cannot read Nanos")
+			}
+			ts.Nanos = nanos
+		}
+	}
+	return nil
+}

app/vlinsert/loki/push_request.proto

@@ -1,38 +0,0 @@
-syntax = "proto3";
-
-// source: https://raw.githubusercontent.com/grafana/loki/main/pkg/push/push.proto
-// Licensed under the Apache License, Version 2.0 (the "License");
-// https://github.com/grafana/loki/blob/main/pkg/push/LICENSE
-
-package logproto;
-
-import "gogoproto/gogo.proto";
-import "google/protobuf/timestamp.proto";
-
-option go_package = "github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/loki";
-
-message PushRequest {
-  repeated StreamAdapter streams = 1 [
-    (gogoproto.jsontag) = "streams",
-    (gogoproto.customtype) = "Stream"
-  ];
-}
-
-message StreamAdapter {
-  string labels = 1 [(gogoproto.jsontag) = "labels"];
-  repeated EntryAdapter entries = 2 [
-    (gogoproto.nullable) = false,
-    (gogoproto.jsontag) = "entries"
-  ];
-  // hash contains the original hash of the stream.
-  uint64 hash = 3 [(gogoproto.jsontag) = "-"];
-}
-
-message EntryAdapter {
-  google.protobuf.Timestamp timestamp = 1 [
-    (gogoproto.stdtime) = true,
-    (gogoproto.nullable) = false,
-    (gogoproto.jsontag) = "ts"
-  ];
-  string line = 2 [(gogoproto.jsontag) = "line"];
-}

app/vlinsert/loki/timestamp.go

@@ -1,110 +0,0 @@
-package loki
-
-// source: https://raw.githubusercontent.com/grafana/loki/main/pkg/push/timestamp.go
-// Licensed under the Apache License, Version 2.0 (the "License");
-// https://github.com/grafana/loki/blob/main/pkg/push/LICENSE
-
-import (
-	"errors"
-	"strconv"
-	"time"
-
-	"github.com/gogo/protobuf/types"
-)
-
-const (
-	// Seconds field of the earliest valid Timestamp.
-	// This is time.Date(1, 1, 1, 0, 0, 0, 0, time.UTC).Unix().
-	minValidSeconds = -62135596800
-	// Seconds field just after the latest valid Timestamp.
-	// This is time.Date(10000, 1, 1, 0, 0, 0, 0, time.UTC).Unix().
-	maxValidSeconds = 253402300800
-)
-
-// validateTimestamp determines whether a Timestamp is valid.
-// A valid timestamp represents a time in the range
-// [0001-01-01, 10000-01-01) and has a Nanos field
-// in the range [0, 1e9).
-//
-// If the Timestamp is valid, validateTimestamp returns nil.
-// Otherwise, it returns an error that describes
-// the problem.
-//
-// Every valid Timestamp can be represented by a time.Time, but the converse is not true.
-func validateTimestamp(ts *types.Timestamp) error {
-	if ts == nil {
-		return errors.New("timestamp: nil Timestamp")
-	}
-	if ts.Seconds < minValidSeconds {
-		return errors.New("timestamp: " + formatTimestamp(ts) + " before 0001-01-01")
-	}
-	if ts.Seconds >= maxValidSeconds {
-		return errors.New("timestamp: " + formatTimestamp(ts) + " after 10000-01-01")
-	}
-	if ts.Nanos < 0 || ts.Nanos >= 1e9 {
-		return errors.New("timestamp: " + formatTimestamp(ts) + ": nanos not in range [0, 1e9)")
-	}
-	return nil
-}
-
-// formatTimestamp is equivalent to fmt.Sprintf("%#v", ts)
-// but avoids the escape incurred by using fmt.Sprintf, eliminating
-// unnecessary heap allocations.
-func formatTimestamp(ts *types.Timestamp) string {
-	if ts == nil {
-		return "nil"
-	}
-
-	seconds := strconv.FormatInt(ts.Seconds, 10)
-	nanos := strconv.FormatInt(int64(ts.Nanos), 10)
-	return "&types.Timestamp{Seconds: " + seconds + ",\nNanos: " + nanos + ",\n}"
-}
-
-func sizeOfStdTime(t time.Time) int {
-	ts, err := timestampProto(t)
-	if err != nil {
-		return 0
-	}
-	return ts.Size()
-}
-
-func stdTimeMarshalTo(t time.Time, data []byte) (int, error) {
-	ts, err := timestampProto(t)
-	if err != nil {
-		return 0, err
-	}
-	return ts.MarshalTo(data)
-}
-
-func stdTimeUnmarshal(t *time.Time, data []byte) error {
-	ts := &types.Timestamp{}
-	if err := ts.Unmarshal(data); err != nil {
-		return err
-	}
-	tt, err := timestampFromProto(ts)
-	if err != nil {
-		return err
-	}
-	*t = tt
-	return nil
-}
-
-func timestampFromProto(ts *types.Timestamp) (time.Time, error) {
-	// Don't return the zero value on error, because corresponds to a valid
-	// timestamp. Instead return whatever time.Unix gives us.
-	var t time.Time
-	if ts == nil {
-		t = time.Unix(0, 0).UTC() // treat nil like the empty Timestamp
-	} else {
-		t = time.Unix(ts.Seconds, int64(ts.Nanos)).UTC()
-	}
-	return t, validateTimestamp(ts)
-}
-
-func timestampProto(t time.Time) (types.Timestamp, error) {
-	ts := types.Timestamp{
-		Seconds: t.Unix(),
-		Nanos:   int32(t.Nanosecond()),
-	}
-	return ts, validateTimestamp(&ts)
-}

app/vlinsert/loki/types.go

@@ -1,481 +0,0 @@
-package loki
-
-// source: https://raw.githubusercontent.com/grafana/loki/main/pkg/push/types.go
-// Licensed under the Apache License, Version 2.0 (the "License");
-// https://github.com/grafana/loki/blob/main/pkg/push/LICENSE
-
-import (
-	"fmt"
-	"io"
-	"time"
-)
-
-// Stream contains a unique labels set as a string and a set of entries for it.
-// We are not using the proto generated version but this custom one so that we
-// can improve serialization see benchmark.
-type Stream struct {
-	Labels  string  `protobuf:"bytes,1,opt,name=labels,proto3" json:"labels"`
-	Entries []Entry `protobuf:"bytes,2,rep,name=entries,proto3,customtype=EntryAdapter" json:"entries"`
-	Hash    uint64  `protobuf:"varint,3,opt,name=hash,proto3" json:"-"`
-}
-
-// Entry is a log entry with a timestamp.
-type Entry struct {
-	Timestamp time.Time `protobuf:"bytes,1,opt,name=timestamp,proto3,stdtime" json:"ts"`
-	Line      string    `protobuf:"bytes,2,opt,name=line,proto3" json:"line"`
-}
-
-// Marshal implements the proto.Marshaler interface.
-func (m *Stream) Marshal() (dAtA []byte, err error) {
-	size := m.Size()
-	dAtA = make([]byte, size)
-	n, err := m.MarshalToSizedBuffer(dAtA[:size])
-	if err != nil {
-		return nil, err
-	}
-	return dAtA[:n], nil
-}
-
-// MarshalTo marshals m to dst.
-func (m *Stream) MarshalTo(dAtA []byte) (int, error) {
-	size := m.Size()
-	return m.MarshalToSizedBuffer(dAtA[:size])
-}
-
-// MarshalToSizedBuffer marshals m to the sized buffer.
-func (m *Stream) MarshalToSizedBuffer(dAtA []byte) (int, error) {
-	i := len(dAtA)
-	_ = i
-	var l int
-	_ = l
-	if m.Hash != 0 {
-		i = encodeVarintPush(dAtA, i, m.Hash)
-		i--
-		dAtA[i] = 0x18
-	}
-	if len(m.Entries) > 0 {
-		for iNdEx := len(m.Entries) - 1; iNdEx >= 0; iNdEx-- {
-			{
-				size, err := m.Entries[iNdEx].MarshalToSizedBuffer(dAtA[:i])
-				if err != nil {
-					return 0, err
-				}
-				i -= size
-				i = encodeVarintPush(dAtA, i, uint64(size))
-			}
-			i--
-			dAtA[i] = 0x12
-		}
-	}
-	if len(m.Labels) > 0 {
-		i -= len(m.Labels)
-		copy(dAtA[i:], m.Labels)
-		i = encodeVarintPush(dAtA, i, uint64(len(m.Labels)))
-		i--
-		dAtA[i] = 0xa
-	}
-	return len(dAtA) - i, nil
-}
-
-// Marshal implements the proto.Marshaler interface.
-func (m *Entry) Marshal() (dAtA []byte, err error) {
-	size := m.Size()
-	dAtA = make([]byte, size)
-	n, err := m.MarshalToSizedBuffer(dAtA[:size])
-	if err != nil {
-		return nil, err
-	}
-	return dAtA[:n], nil
-}
-
-// MarshalTo marshals m to dst.
-func (m *Entry) MarshalTo(dAtA []byte) (int, error) {
-	size := m.Size()
-	return m.MarshalToSizedBuffer(dAtA[:size])
-}
-
-// MarshalToSizedBuffer marshals m to the sized buffer.
-func (m *Entry) MarshalToSizedBuffer(dAtA []byte) (int, error) {
-	i := len(dAtA)
-	_ = i
-	var l int
-	_ = l
-	if len(m.Line) > 0 {
-		i -= len(m.Line)
-		copy(dAtA[i:], m.Line)
-		i = encodeVarintPush(dAtA, i, uint64(len(m.Line)))
-		i--
-		dAtA[i] = 0x12
-	}
-	n7, err7 := stdTimeMarshalTo(m.Timestamp, dAtA[i-sizeOfStdTime(m.Timestamp):])
-	if err7 != nil {
-		return 0, err7
-	}
-	i -= n7
-	i = encodeVarintPush(dAtA, i, uint64(n7))
-	i--
-	dAtA[i] = 0xa
-	return len(dAtA) - i, nil
-}
-
-// Unmarshal unmarshals the given data into m.
-func (m *Stream) Unmarshal(dAtA []byte) error {
-	l := len(dAtA)
-	iNdEx := 0
-	for iNdEx < l {
-		preIndex := iNdEx
-		var wire uint64
-		for shift := uint(0); ; shift += 7 {
-			if shift >= 64 {
-				return ErrIntOverflowPush
-			}
-			if iNdEx >= l {
-				return io.ErrUnexpectedEOF
-			}
-			b := dAtA[iNdEx]
-			iNdEx++
-			wire |= uint64(b&0x7F) << shift
-			if b < 0x80 {
-				break
-			}
-		}
-		fieldNum := int32(wire >> 3)
-		wireType := int(wire & 0x7)
-		if wireType == 4 {
-			return fmt.Errorf("proto: StreamAdapter: wiretype end group for non-group")
-		}
-		if fieldNum <= 0 {
-			return fmt.Errorf("proto: StreamAdapter: illegal tag %d (wire type %d)", fieldNum, wire)
-		}
-		switch fieldNum {
-		case 1:
-			if wireType != 2 {
-				return fmt.Errorf("proto: wrong wireType = %d for field Labels", wireType)
-			}
-			var stringLen uint64
-			for shift := uint(0); ; shift += 7 {
-				if shift >= 64 {
-					return ErrIntOverflowPush
-				}
-				if iNdEx >= l {
-					return io.ErrUnexpectedEOF
-				}
-				b := dAtA[iNdEx]
-				iNdEx++
-				stringLen |= uint64(b&0x7F) << shift
-				if b < 0x80 {
-					break
-				}
-			}
-			intStringLen := int(stringLen)
-			if intStringLen < 0 {
-				return ErrInvalidLengthPush
-			}
-			postIndex := iNdEx + intStringLen
-			if postIndex < 0 {
-				return ErrInvalidLengthPush
-			}
-			if postIndex > l {
-				return io.ErrUnexpectedEOF
-			}
-			m.Labels = string(dAtA[iNdEx:postIndex])
-			iNdEx = postIndex
-		case 2:
-			if wireType != 2 {
-				return fmt.Errorf("proto: wrong wireType = %d for field Entries", wireType)
-			}
-			var msglen int
-			for shift := uint(0); ; shift += 7 {
-				if shift >= 64 {
-					return ErrIntOverflowPush
-				}
-				if iNdEx >= l {
-					return io.ErrUnexpectedEOF
-				}
-				b := dAtA[iNdEx]
-				iNdEx++
-				msglen |= int(b&0x7F) << shift
-				if b < 0x80 {
-					break
-				}
-			}
-			if msglen < 0 {
-				return ErrInvalidLengthPush
-			}
-			postIndex := iNdEx + msglen
-			if postIndex < 0 {
-				return ErrInvalidLengthPush
-			}
-			if postIndex > l {
-				return io.ErrUnexpectedEOF
-			}
-			m.Entries = append(m.Entries, Entry{})
-			if err := m.Entries[len(m.Entries)-1].Unmarshal(dAtA[iNdEx:postIndex]); err != nil {
-				return err
-			}
-			iNdEx = postIndex
-		case 3:
-			if wireType != 0 {
-				return fmt.Errorf("proto: wrong wireType = %d for field Hash", wireType)
-			}
-			m.Hash = 0
-			for shift := uint(0); ; shift += 7 {
-				if shift >= 64 {
-					return ErrIntOverflowPush
-				}
-				if iNdEx >= l {
-					return io.ErrUnexpectedEOF
-				}
-				b := dAtA[iNdEx]
-				iNdEx++
-				m.Hash |= uint64(b&0x7F) << shift
-				if b < 0x80 {
-					break
-				}
-			}
-		default:
-			iNdEx = preIndex
-			skippy, err := skipPush(dAtA[iNdEx:])
-			if err != nil {
-				return err
-			}
-			if skippy < 0 {
-				return ErrInvalidLengthPush
-			}
-			if (iNdEx + skippy) < 0 {
-				return ErrInvalidLengthPush
-			}
-			if (iNdEx + skippy) > l {
-				return io.ErrUnexpectedEOF
-			}
-			iNdEx += skippy
-		}
-	}
-
-	if iNdEx > l {
-		return io.ErrUnexpectedEOF
-	}
-	return nil
-}
-
-// Unmarshal unmarshals the given data into m.
-func (m *Entry) Unmarshal(dAtA []byte) error {
-	l := len(dAtA)
-	iNdEx := 0
-	for iNdEx < l {
-		preIndex := iNdEx
-		var wire uint64
-		for shift := uint(0); ; shift += 7 {
-			if shift >= 64 {
-				return ErrIntOverflowPush
-			}
-			if iNdEx >= l {
-				return io.ErrUnexpectedEOF
-			}
-			b := dAtA[iNdEx]
-			iNdEx++
-			wire |= uint64(b&0x7F) << shift
-			if b < 0x80 {
-				break
-			}
-		}
-		fieldNum := int32(wire >> 3)
-		wireType := int(wire & 0x7)
-		if wireType == 4 {
-			return fmt.Errorf("proto: EntryAdapter: wiretype end group for non-group")
-		}
-		if fieldNum <= 0 {
-			return fmt.Errorf("proto: EntryAdapter: illegal tag %d (wire type %d)", fieldNum, wire)
-		}
-		switch fieldNum {
-		case 1:
-			if wireType != 2 {
-				return fmt.Errorf("proto: wrong wireType = %d for field Timestamp", wireType)
-			}
-			var msglen int
-			for shift := uint(0); ; shift += 7 {
-				if shift >= 64 {
-					return ErrIntOverflowPush
-				}
-				if iNdEx >= l {
-					return io.ErrUnexpectedEOF
-				}
-				b := dAtA[iNdEx]
-				iNdEx++
-				msglen |= int(b&0x7F) << shift
-				if b < 0x80 {
-					break
-				}
-			}
-			if msglen < 0 {
-				return ErrInvalidLengthPush
-			}
-			postIndex := iNdEx + msglen
-			if postIndex < 0 {
-				return ErrInvalidLengthPush
-			}
-			if postIndex > l {
-				return io.ErrUnexpectedEOF
-			}
-			if err := stdTimeUnmarshal(&m.Timestamp, dAtA[iNdEx:postIndex]); err != nil {
-				return err
-			}
-			iNdEx = postIndex
-		case 2:
-			if wireType != 2 {
-				return fmt.Errorf("proto: wrong wireType = %d for field Line", wireType)
-			}
-			var stringLen uint64
-			for shift := uint(0); ; shift += 7 {
-				if shift >= 64 {
-					return ErrIntOverflowPush
-				}
-				if iNdEx >= l {
-					return io.ErrUnexpectedEOF
-				}
-				b := dAtA[iNdEx]
-				iNdEx++
-				stringLen |= uint64(b&0x7F) << shift
-				if b < 0x80 {
-					break
-				}
-			}
-			intStringLen := int(stringLen)
-			if intStringLen < 0 {
-				return ErrInvalidLengthPush
-			}
-			postIndex := iNdEx + intStringLen
-			if postIndex < 0 {
-				return ErrInvalidLengthPush
-			}
-			if postIndex > l {
-				return io.ErrUnexpectedEOF
-			}
-			m.Line = string(dAtA[iNdEx:postIndex])
-			iNdEx = postIndex
-		default:
-			iNdEx = preIndex
-			skippy, err := skipPush(dAtA[iNdEx:])
-			if err != nil {
-				return err
-			}
-			if skippy < 0 {
-				return ErrInvalidLengthPush
-			}
-			if (iNdEx + skippy) < 0 {
-				return ErrInvalidLengthPush
-			}
-			if (iNdEx + skippy) > l {
-				return io.ErrUnexpectedEOF
-			}
-			iNdEx += skippy
-		}
-	}
-
-	if iNdEx > l {
-		return io.ErrUnexpectedEOF
-	}
-	return nil
-}
-
-// Size returns the size of the serialized Stream.
-func (m *Stream) Size() (n int) {
-	if m == nil {
-		return 0
-	}
-	var l int
-	_ = l
-	l = len(m.Labels)
-	if l > 0 {
-		n += 1 + l + sovPush(uint64(l))
-	}
-	if len(m.Entries) > 0 {
-		for _, e := range m.Entries {
-			l = e.Size()
-			n += 1 + l + sovPush(uint64(l))
-		}
-	}
-	if m.Hash != 0 {
-		n += 1 + sovPush(m.Hash)
-	}
-	return n
-}
-
-// Size returns the size of the serialized Entry
-func (m *Entry) Size() (n int) {
-	if m == nil {
-		return 0
-	}
-	var l int
-	_ = l
-	l = sizeOfStdTime(m.Timestamp)
-	n += 1 + l + sovPush(uint64(l))
-	l = len(m.Line)
-	if l > 0 {
-		n += 1 + l + sovPush(uint64(l))
-	}
-	return n
-}
-
-// Equal returns true if the two Streams are equal.
-func (m *Stream) Equal(that interface{}) bool {
-	if that == nil {
-		return m == nil
-	}
-
-	that1, ok := that.(*Stream)
-	if !ok {
-		that2, ok := that.(Stream)
-		if ok {
-			that1 = &that2
-		} else {
-			return false
-		}
-	}
-	if that1 == nil {
-		return m == nil
-	} else if m == nil {
-		return false
-	}
-	if m.Labels != that1.Labels {
-		return false
-	}
-	if len(m.Entries) != len(that1.Entries) {
-		return false
-	}
-	for i := range m.Entries {
-		if !m.Entries[i].Equal(that1.Entries[i]) {
-			return false
-		}
-	}
-	return m.Hash == that1.Hash
-}
-
-// Equal returns true if the two Entries are equal.
-func (m *Entry) Equal(that interface{}) bool {
-	if that == nil {
-		return m == nil
-	}
-
-	that1, ok := that.(*Entry)
-	if !ok {
-		that2, ok := that.(Entry)
-		if ok {
-			that1 = &that2
-		} else {
-			return false
-		}
-	}
-	if that1 == nil {
-		return m == nil
-	} else if m == nil {
-		return false
-	}
-	if !m.Timestamp.Equal(that1.Timestamp) {
-		return false
-	}
-	if m.Line != that1.Line {
-		return false
-	}
-	return true
-}

app/vlinsert/syslog/syslog.go

@@ -46,24 +46,29 @@ var (
 		"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
 
 	tlsEnable = flagutil.NewArrayBool("syslog.tls", "Whether to enable TLS for receiving syslog messages at the corresponding -syslog.listenAddr.tcp. "+
-		"The corresponding -syslog.tlsCertFile and -syslog.tlsKeyFile must be set if -syslog.tls is set. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
+		"The corresponding -syslog.tlsCertFile and -syslog.tlsKeyFile must be set if -syslog.tls is set. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#security")
 	tlsCertFile = flagutil.NewArrayString("syslog.tlsCertFile", "Path to file with TLS certificate for the corresponding -syslog.listenAddr.tcp if the corresponding -syslog.tls is set. "+
 		"Prefer ECDSA certs instead of RSA certs as RSA certs are slower. The provided certificate file is automatically re-read every second, so it can be dynamically updated. "+
-		"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
+		"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#security")
 	tlsKeyFile = flagutil.NewArrayString("syslog.tlsKeyFile", "Path to file with TLS key for the corresponding -syslog.listenAddr.tcp if the corresponding -syslog.tls is set. "+
 		"The provided key file is automatically re-read every second, so it can be dynamically updated. "+
-		"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
+		"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#security")
 	tlsCipherSuites = flagutil.NewArrayString("syslog.tlsCipherSuites", "Optional list of TLS cipher suites for -syslog.listenAddr.tcp if -syslog.tls is set. "+
 		"See the list of supported cipher suites at https://pkg.go.dev/crypto/tls#pkg-constants . "+
-		"See also https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
+		"See also https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#security")
 	tlsMinVersion = flag.String("syslog.tlsMinVersion", "TLS13", "The minimum TLS version to use for -syslog.listenAddr.tcp if -syslog.tls is set. "+
 		"Supported values: TLS10, TLS11, TLS12, TLS13. "+
-		"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
+		"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#security")
 
 	compressMethodTCP = flagutil.NewArrayString("syslog.compressMethod.tcp", "Compression method for syslog messages received at the corresponding -syslog.listenAddr.tcp. "+
-		"Supported values: none, gzip, deflate. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
+		"Supported values: none, gzip, deflate. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#compression")
 	compressMethodUDP = flagutil.NewArrayString("syslog.compressMethod.udp", "Compression method for syslog messages received at the corresponding -syslog.listenAddr.udp. "+
-		"Supported values: none, gzip, deflate. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
+		"Supported values: none, gzip, deflate. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#compression")
+
+	useLocalTimestampTCP = flagutil.NewArrayBool("syslog.useLocalTimestamp.tcp", "Whether to use local timestamp instead of the original timestamp for the ingested syslog messages "+
+		"at the corresponding -syslog.listenAddr.tcp. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#log-timestamps")
+	useLocalTimestampUDP = flagutil.NewArrayBool("syslog.useLocalTimestamp.udp", "Whether to use local timestamp instead of the original timestamp for the ingested syslog messages "+
+		"at the corresponding -syslog.listenAddr.udp. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#log-timestamps")
 )
 
 // MustInit initializes syslog parser at the given -syslog.listenAddr.tcp and -syslog.listenAddr.udp ports
@@ -154,17 +159,21 @@ func runUDPListener(addr string, argIdx int) {
 	compressMethod := compressMethodUDP.GetOptionalArg(argIdx)
 	checkCompressMethod(compressMethod, addr, "udp")
 
+	useLocalTimestamp := useLocalTimestampUDP.GetOptionalArg(argIdx)
+
 	doneCh := make(chan struct{})
 	go func() {
-		serveUDP(ln, tenantID, compressMethod)
+		serveUDP(ln, tenantID, compressMethod, useLocalTimestamp)
 		close(doneCh)
 	}()
 
+	logger.Infof("started accepting syslog messages at -syslog.listenAddr.udp=%q", addr)
 	<-workersStopCh
 	if err := ln.Close(); err != nil {
 		logger.Fatalf("syslog: cannot close UDP listener at %s: %s", addr, err)
 	}
 	<-doneCh
+	logger.Infof("finished accepting syslog messages at -syslog.listenAddr.udp=%q", addr)
 }
 
 func runTCPListener(addr string, argIdx int) {
@@ -193,17 +202,21 @@ func runTCPListener(addr string, argIdx int) {
 	compressMethod := compressMethodTCP.GetOptionalArg(argIdx)
 	checkCompressMethod(compressMethod, addr, "tcp")
 
+	useLocalTimestamp := useLocalTimestampTCP.GetOptionalArg(argIdx)
+
 	doneCh := make(chan struct{})
 	go func() {
-		serveTCP(ln, tenantID, compressMethod)
+		serveTCP(ln, tenantID, compressMethod, useLocalTimestamp)
 		close(doneCh)
 	}()
 
+	logger.Infof("started accepting syslog messages at -syslog.listenAddr.tcp=%q", addr)
 	<-workersStopCh
 	if err := ln.Close(); err != nil {
 		logger.Fatalf("syslog: cannot close TCP listener at %s: %s", addr, err)
 	}
 	<-doneCh
+	logger.Infof("finished accepting syslog messages at -syslog.listenAddr.tcp=%q", addr)
 }
 
 func checkCompressMethod(compressMethod, addr, protocol string) {
@@ -215,7 +228,7 @@ func checkCompressMethod(compressMethod, addr, protocol string) {
 	}
 }
 
-func serveUDP(ln net.PacketConn, tenantID logstorage.TenantID, compressMethod string) {
+func serveUDP(ln net.PacketConn, tenantID logstorage.TenantID, compressMethod string, useLocalTimestamp bool) {
 	gomaxprocs := cgroup.AvailableCPUs()
 	var wg sync.WaitGroup
 	localAddr := ln.LocalAddr()
@@ -248,7 +261,7 @@ func serveUDP(ln net.PacketConn, tenantID logstorage.TenantID, compressMethod st
 				}
 				bb.B = bb.B[:n]
 				udpRequestsTotal.Inc()
-				if err := processStream(bb.NewReader(), compressMethod, cp); err != nil {
+				if err := processStream(bb.NewReader(), compressMethod, useLocalTimestamp, cp); err != nil {
 					logger.Errorf("syslog: cannot process UDP data from %s at %s: %s", remoteAddr, localAddr, err)
 				}
 			}
@@ -257,7 +270,7 @@ func serveUDP(ln net.PacketConn, tenantID logstorage.TenantID, compressMethod st
 	wg.Wait()
 }
 
-func serveTCP(ln net.Listener, tenantID logstorage.TenantID, compressMethod string) {
+func serveTCP(ln net.Listener, tenantID logstorage.TenantID, compressMethod string, useLocalTimestamp bool) {
 	var cm ingestserver.ConnsMap
 	cm.Init("syslog")
 
@@ -288,7 +301,7 @@ func serveTCP(ln net.Listener, tenantID logstorage.TenantID, compressMethod stri
 		wg.Add(1)
 		go func() {
 			cp := insertutils.GetCommonParamsForSyslog(tenantID)
-			if err := processStream(c, compressMethod, cp); err != nil {
+			if err := processStream(c, compressMethod, useLocalTimestamp, cp); err != nil {
 				logger.Errorf("syslog: cannot process TCP data at %q: %s", addr, err)
 			}
 
@@ -303,19 +316,19 @@ func serveTCP(ln net.Listener, tenantID logstorage.TenantID, compressMethod stri
 }
 
 // processStream parses a stream of syslog messages from r and ingests them into vlstorage.
-func processStream(r io.Reader, compressMethod string, cp *insertutils.CommonParams) error {
+func processStream(r io.Reader, compressMethod string, useLocalTimestamp bool, cp *insertutils.CommonParams) error {
 	if err := vlstorage.CanWriteData(); err != nil {
 		return err
 	}
 
 	lmp := cp.NewLogMessageProcessor()
-	err := processStreamInternal(r, compressMethod, lmp)
+	err := processStreamInternal(r, compressMethod, useLocalTimestamp, lmp)
 	lmp.MustClose()
 
 	return err
 }
 
-func processStreamInternal(r io.Reader, compressMethod string, lmp insertutils.LogMessageProcessor) error {
+func processStreamInternal(r io.Reader, compressMethod string, useLocalTimestamp bool, lmp insertutils.LogMessageProcessor) error {
 	switch compressMethod {
 	case "", "none":
 	case "gzip":
@@ -334,7 +347,7 @@ func processStreamInternal(r io.Reader, compressMethod string, lmp insertutils.L
 		logger.Panicf("BUG: unsupported compressMethod=%q; supported values: none, gzip, deflate", compressMethod)
 	}
 
-	err := processUncompressedStream(r, lmp)
+	err := processUncompressedStream(r, useLocalTimestamp, lmp)
 
 	switch compressMethod {
 	case "gzip":
@@ -348,7 +361,7 @@ func processStreamInternal(r io.Reader, compressMethod string, lmp insertutils.L
 	return err
 }
 
-func processUncompressedStream(r io.Reader, lmp insertutils.LogMessageProcessor) error {
+func processUncompressedStream(r io.Reader, useLocalTimestamp bool, lmp insertutils.LogMessageProcessor) error {
 	wcr := writeconcurrencylimiter.GetReader(r)
 	defer writeconcurrencylimiter.PutReader(wcr)
 
@@ -364,7 +377,7 @@ func processUncompressedStream(r io.Reader, lmp insertutils.LogMessageProcessor)
 		}
 
 		currentYear := int(globalCurrentYear.Load())
-		err := processLine(slr.line, currentYear, globalTimezone, lmp)
+		err := processLine(slr.line, currentYear, globalTimezone, useLocalTimestamp, lmp)
 		if err != nil {
 			errorsTotal.Inc()
 			return fmt.Errorf("cannot read line #%d: %s", n, err)
@@ -405,6 +418,7 @@ func (slr *syslogLineReader) nextLine() bool {
 		return false
 	}
 
+again:
 	prefix, err := slr.br.ReadSlice(' ')
 	if err != nil {
 		if err != io.EOF {
@@ -420,6 +434,10 @@ func (slr *syslogLineReader) nextLine() bool {
 	for len(prefix) > 0 && prefix[0] == '\n' {
 		prefix = prefix[1:]
 	}
+	if len(prefix) == 0 {
+		// An empty prefix or a prefix with empty lines - try reading yet another prefix.
+		goto again
+	}
 
 	if prefix[0] >= '0' && prefix[0] <= '9' {
 		// This is octet-counting method. See https://www.ietf.org/archive/id/draft-gerhards-syslog-plain-tcp-07.html#msgxfer
@@ -481,13 +499,20 @@ func putSyslogLineReader(slr *syslogLineReader) {
 
 var syslogLineReaderPool sync.Pool
 
-func processLine(line []byte, currentYear int, timezone *time.Location, lmp insertutils.LogMessageProcessor) error {
+func processLine(line []byte, currentYear int, timezone *time.Location, useLocalTimestamp bool, lmp insertutils.LogMessageProcessor) error {
 	p := logstorage.GetSyslogParser(currentYear, timezone)
 	lineStr := bytesutil.ToUnsafeString(line)
 	p.Parse(lineStr)
-	ts, err := insertutils.ExtractTimestampRFC3339NanoFromFields("timestamp", p.Fields)
-	if err != nil {
-		return fmt.Errorf("cannot get timestamp from syslog line %q: %w", line, err)
+
+	var ts int64
+	if useLocalTimestamp {
+		ts = time.Now().UnixNano()
+	} else {
+		nsecs, err := insertutils.ExtractTimestampRFC3339NanoFromFields("timestamp", p.Fields)
+		if err != nil {
+			return fmt.Errorf("cannot get timestamp from syslog line %q: %w", line, err)
+		}
+		ts = nsecs
 	}
 	logstorage.RenameField(p.Fields, "message", "_msg")
 	lmp.AddRow(ts, p.Fields)

app/vlinsert/syslog/syslog_test.go

@@ -30,10 +30,19 @@ func TestSyslogLineReader_Success(t *testing.T) {
 	}
 
 	f("", nil)
+	f("\n", nil)
+	f("\n\n\n", nil)
+
 	f("foobar", []string{"foobar"})
+	f("foobar\n", []string{"foobar\n"})
+	f("\n\nfoo\n\nbar\n\n", []string{"foo\n\nbar\n\n"})
+
+	f(`Jun  3 12:08:33 abcd systemd: Starting Update the local ESM caches...`, []string{"Jun  3 12:08:33 abcd systemd: Starting Update the local ESM caches..."})
 
 	f(`Jun  3 12:08:33 abcd systemd: Starting Update the local ESM caches...
+
 48 <165>Jun  4 12:08:33 abcd systemd[345]: abc defg<123>1 2023-06-03T17:42:12.345Z mymachine.example.com appname 12345 ID47 [exampleSDID@32473 iut="3" eventSource="Application 123 = ] 56" eventID="11211"] This is a test message with structured data.
+
 `, []string{
 		"Jun  3 12:08:33 abcd systemd: Starting Update the local ESM caches...",
 		"<165>Jun  4 12:08:33 abcd systemd[345]: abc defg",
@@ -77,7 +86,7 @@ func TestProcessStreamInternal_Success(t *testing.T) {
 
 		tlp := &insertutils.TestLogMessageProcessor{}
 		r := bytes.NewBufferString(data)
-		if err := processStreamInternal(r, "", tlp); err != nil {
+		if err := processStreamInternal(r, "", false, tlp); err != nil {
 			t.Fatalf("unexpected error: %s", err)
 		}
 		if err := tlp.Verify(rowsExpected, timestampsExpected, resultExpected); err != nil {
@@ -94,7 +103,7 @@ func TestProcessStreamInternal_Success(t *testing.T) {
 	timestampsExpected := []int64{1685794113000000000, 1685880513000000000, 1685814132345000000}
 	resultExpected := `{"format":"rfc3164","timestamp":"","hostname":"abcd","app_name":"systemd","_msg":"Starting Update the local ESM caches..."}
 {"priority":"165","facility":"20","severity":"5","format":"rfc3164","timestamp":"","hostname":"abcd","app_name":"systemd","proc_id":"345","_msg":"abc defg"}
-{"priority":"123","facility":"15","severity":"3","format":"rfc5424","timestamp":"","hostname":"mymachine.example.com","app_name":"appname","proc_id":"12345","msg_id":"ID47","exampleSDID@32473":"iut=\"3\" eventSource=\"Application 123 = ] 56\" eventID=\"11211\"","_msg":"This is a test message with structured data."}`
+{"priority":"123","facility":"15","severity":"3","format":"rfc5424","timestamp":"","hostname":"mymachine.example.com","app_name":"appname","proc_id":"12345","msg_id":"ID47","exampleSDID@32473.iut":"3","exampleSDID@32473.eventSource":"Application 123 = ] 56","exampleSDID@32473.eventID":"11211","_msg":"This is a test message with structured data."}`
 	f(data, currentYear, rowsExpected, timestampsExpected, resultExpected)
 }
 
@@ -107,7 +116,7 @@ func TestProcessStreamInternal_Failure(t *testing.T) {
 
 		tlp := &insertutils.TestLogMessageProcessor{}
 		r := bytes.NewBufferString(data)
-		if err := processStreamInternal(r, "", tlp); err == nil {
+		if err := processStreamInternal(r, "", false, tlp); err == nil {
 			t.Fatalf("expecting non-nil error")
 		}
 	}

app/vlselect/logsql/hits_response.qtpl

@@ -44,7 +44,7 @@
 		hs := m[k]
 		hs.sort()
 		timestamps := hs.timestamps
-		values := hs.values
+		hits := hs.hits
 	%}
 	"fields":{%s= k %},
 	"timestamps":[
@@ -56,13 +56,14 @@
 		{% endif %}
 	],
 	"values":[
-		{% if len(values) > 0 %}
-			{%s= values[0] %}
-			{% for _, v := range values[1:] %}
-				,{%s= v %}
+		{% if len(hits) > 0 %}
+			{%dul= hits[0] %}
+			{% for _, v := range hits[1:] %}
+				,{%dul= v %}
 			{% endfor %}
 		{% endif %}
-	]
+	],
+	"total":{%dul= hs.hitsTotal %}
 }
 {% endfunc %}
 

app/vlselect/logsql/hits_response.qtpl.go

@@ -149,7 +149,7 @@ func streamhitsSeriesLine(qw422016 *qt422016.Writer, m map[string]*hitsSeries, k
 	hs := m[k]
 	hs.sort()
 	timestamps := hs.timestamps
-	values := hs.values
+	hits := hs.hits
 
 //line app/vlselect/logsql/hits_response.qtpl:48
 	qw422016.N().S(`"fields":`)
@@ -174,46 +174,50 @@ func streamhitsSeriesLine(qw422016 *qt422016.Writer, m map[string]*hitsSeries, k
 //line app/vlselect/logsql/hits_response.qtpl:56
 	qw422016.N().S(`],"values":[`)
 //line app/vlselect/logsql/hits_response.qtpl:59
-	if len(values) > 0 {
+	if len(hits) > 0 {
 //line app/vlselect/logsql/hits_response.qtpl:60
-		qw422016.N().S(values[0])
+		qw422016.N().DUL(hits[0])
 //line app/vlselect/logsql/hits_response.qtpl:61
-		for _, v := range values[1:] {
+		for _, v := range hits[1:] {
 //line app/vlselect/logsql/hits_response.qtpl:61
 			qw422016.N().S(`,`)
 //line app/vlselect/logsql/hits_response.qtpl:62
-			qw422016.N().S(v)
+			qw422016.N().DUL(v)
 //line app/vlselect/logsql/hits_response.qtpl:63
 		}
 //line app/vlselect/logsql/hits_response.qtpl:64
 	}
 //line app/vlselect/logsql/hits_response.qtpl:64
-	qw422016.N().S(`]}`)
-//line app/vlselect/logsql/hits_response.qtpl:67
+	qw422016.N().S(`],"total":`)
+//line app/vlselect/logsql/hits_response.qtpl:66
+	qw422016.N().DUL(hs.hitsTotal)
+//line app/vlselect/logsql/hits_response.qtpl:66
+	qw422016.N().S(`}`)
+//line app/vlselect/logsql/hits_response.qtpl:68
 }
 
-//line app/vlselect/logsql/hits_response.qtpl:67
+//line app/vlselect/logsql/hits_response.qtpl:68
 func writehitsSeriesLine(qq422016 qtio422016.Writer, m map[string]*hitsSeries, k string) {
-//line app/vlselect/logsql/hits_response.qtpl:67
+//line app/vlselect/logsql/hits_response.qtpl:68
 	qw422016 := qt422016.AcquireWriter(qq422016)
-//line app/vlselect/logsql/hits_response.qtpl:67
+//line app/vlselect/logsql/hits_response.qtpl:68
 	streamhitsSeriesLine(qw422016, m, k)
-//line app/vlselect/logsql/hits_response.qtpl:67
+//line app/vlselect/logsql/hits_response.qtpl:68
 	qt422016.ReleaseWriter(qw422016)
-//line app/vlselect/logsql/hits_response.qtpl:67
+//line app/vlselect/logsql/hits_response.qtpl:68
 }
 
-//line app/vlselect/logsql/hits_response.qtpl:67
+//line app/vlselect/logsql/hits_response.qtpl:68
 func hitsSeriesLine(m map[string]*hitsSeries, k string) string {
-//line app/vlselect/logsql/hits_response.qtpl:67
+//line app/vlselect/logsql/hits_response.qtpl:68
 	qb422016 := qt422016.AcquireByteBuffer()
-//line app/vlselect/logsql/hits_response.qtpl:67
+//line app/vlselect/logsql/hits_response.qtpl:68
 	writehitsSeriesLine(qb422016, m, k)
-//line app/vlselect/logsql/hits_response.qtpl:67
+//line app/vlselect/logsql/hits_response.qtpl:68
 	qs422016 := string(qb422016.B)
-//line app/vlselect/logsql/hits_response.qtpl:67
+//line app/vlselect/logsql/hits_response.qtpl:68
 	qt422016.ReleaseByteBuffer(qb422016)
-//line app/vlselect/logsql/hits_response.qtpl:67
+//line app/vlselect/logsql/hits_response.qtpl:68
 	return qs422016
-//line app/vlselect/logsql/hits_response.qtpl:67
+//line app/vlselect/logsql/hits_response.qtpl:68
 }

app/vlselect/logsql/logsql.go

@@ -6,14 +6,18 @@ import (
 	"math"
 	"net/http"
 	"sort"
+	"strconv"
 	"strings"
 	"sync"
 	"time"
 
+	"github.com/VictoriaMetrics/metrics"
+
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputils"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promutils"
 )
@@ -56,9 +60,20 @@ func ProcessHitsRequest(ctx context.Context, w http.ResponseWriter, r *http.Requ
 	// Obtain field entries
 	fields := r.Form["field"]
 
-	// Prepare the query
-	q.AddCountByTimePipe(int64(step), int64(offset), fields)
+	// Obtain limit on the number of top fields entries.
+	fieldsLimit, err := httputils.GetInt(r, "fields_limit")
+	if err != nil {
+		httpserver.Errorf(w, r, "%s", err)
+		return
+	}
+	if fieldsLimit < 0 {
+		fieldsLimit = 0
+	}
+
+	// Prepare the query for hits count.
 	q.Optimize()
+	q.DropAllPipes()
+	q.AddCountByTimePipe(int64(step), int64(offset), fields)
 
 	var mLock sync.Mutex
 	m := make(map[string]*hitsSeries)
@@ -75,6 +90,10 @@ func ProcessHitsRequest(ctx context.Context, w http.ResponseWriter, r *http.Requ
 		for i := range timestamps {
 			timestampStr := strings.Clone(timestampValues[i])
 			hitsStr := strings.Clone(hitsValues[i])
+			hits, err := strconv.ParseUint(hitsStr, 10, 64)
+			if err != nil {
+				logger.Panicf("BUG: cannot parse hitsStr=%q: %s", hitsStr, err)
+			}
 
 			bb.Reset()
 			WriteFieldsForHits(bb, columns, i)
@@ -87,7 +106,8 @@ func ProcessHitsRequest(ctx context.Context, w http.ResponseWriter, r *http.Requ
 				m[k] = hs
 			}
 			hs.timestamps = append(hs.timestamps, timestampStr)
-			hs.values = append(hs.values, hitsStr)
+			hs.hits = append(hs.hits, hits)
+			hs.hitsTotal += hits
 			mLock.Unlock()
 		}
 		blockResultPool.Put(bb)
@@ -99,14 +119,59 @@ func ProcessHitsRequest(ctx context.Context, w http.ResponseWriter, r *http.Requ
 		return
 	}
 
+	m = getTopHitsSeries(m, fieldsLimit)
+
 	// Write response
 	w.Header().Set("Content-Type", "application/json")
 	WriteHitsSeries(w, m)
 }
 
+func getTopHitsSeries(m map[string]*hitsSeries, fieldsLimit int) map[string]*hitsSeries {
+	if fieldsLimit <= 0 || fieldsLimit >= len(m) {
+		return m
+	}
+
+	type fieldsHits struct {
+		fieldsStr string
+		hs        *hitsSeries
+	}
+	a := make([]fieldsHits, 0, len(m))
+	for fieldsStr, hs := range m {
+		a = append(a, fieldsHits{
+			fieldsStr: fieldsStr,
+			hs:        hs,
+		})
+	}
+	sort.Slice(a, func(i, j int) bool {
+		return a[i].hs.hitsTotal > a[j].hs.hitsTotal
+	})
+
+	hitsOther := make(map[string]uint64)
+	for _, x := range a[fieldsLimit:] {
+		for i, timestampStr := range x.hs.timestamps {
+			hitsOther[timestampStr] += x.hs.hits[i]
+		}
+	}
+	var hsOther hitsSeries
+	for timestampStr, hits := range hitsOther {
+		hsOther.timestamps = append(hsOther.timestamps, timestampStr)
+		hsOther.hits = append(hsOther.hits, hits)
+		hsOther.hitsTotal += hits
+	}
+
+	mNew := make(map[string]*hitsSeries, fieldsLimit+1)
+	for _, x := range a[:fieldsLimit] {
+		mNew[x.fieldsStr] = x.hs
+	}
+	mNew["{}"] = &hsOther
+
+	return mNew
+}
+
 type hitsSeries struct {
+	hitsTotal  uint64
 	timestamps []string
-	values     []string
+	hits       []uint64
 }
 
 func (hs *hitsSeries) sort() {
@@ -119,7 +184,7 @@ func (hs *hitsSeries) Len() int {
 
 func (hs *hitsSeries) Swap(i, j int) {
 	hs.timestamps[i], hs.timestamps[j] = hs.timestamps[j], hs.timestamps[i]
-	hs.values[i], hs.values[j] = hs.values[j], hs.values[i]
+	hs.hits[i], hs.hits[j] = hs.hits[j], hs.hits[i]
 }
 
 func (hs *hitsSeries) Less(i, j int) bool {
@@ -250,6 +315,38 @@ func ProcessStreamFieldValuesRequest(ctx context.Context, w http.ResponseWriter,
 	WriteValuesWithHitsJSON(w, values)
 }
 
+// ProcessStreamIDsRequest processes /select/logsql/stream_ids request.
+//
+// See https://docs.victoriametrics.com/victorialogs/querying/#querying-stream_ids
+func ProcessStreamIDsRequest(ctx context.Context, w http.ResponseWriter, r *http.Request) {
+	q, tenantIDs, err := parseCommonArgs(r)
+	if err != nil {
+		httpserver.Errorf(w, r, "%s", err)
+		return
+	}
+
+	// Parse limit query arg
+	limit, err := httputils.GetInt(r, "limit")
+	if err != nil {
+		httpserver.Errorf(w, r, "%s", err)
+		return
+	}
+	if limit < 0 {
+		limit = 0
+	}
+
+	// Obtain streamIDs for the given query
+	q.Optimize()
+	streamIDs, err := vlstorage.GetStreamIDs(ctx, tenantIDs, q, uint64(limit))
+	if err != nil {
+		httpserver.Errorf(w, r, "cannot obtain stream_ids: %s", err)
+	}
+
+	// Write results
+	w.Header().Set("Content-Type", "application/json")
+	WriteValuesWithHitsJSON(w, streamIDs)
+}
+
 // ProcessStreamsRequest processes /select/logsql/streams request.
 //
 // See https://docs.victoriametrics.com/victorialogs/querying/#querying-streams
@@ -282,6 +379,187 @@ func ProcessStreamsRequest(ctx context.Context, w http.ResponseWriter, r *http.R
 	WriteValuesWithHitsJSON(w, streams)
 }
 
+// ProcessLiveTailRequest processes live tailing request to /select/logsq/tail
+func ProcessLiveTailRequest(ctx context.Context, w http.ResponseWriter, r *http.Request) {
+	liveTailRequests.Inc()
+	defer liveTailRequests.Dec()
+
+	q, tenantIDs, err := parseCommonArgs(r)
+	if err != nil {
+		httpserver.Errorf(w, r, "%s", err)
+		return
+	}
+	if !q.CanLiveTail() {
+		httpserver.Errorf(w, r, "the query [%s] cannot be used in live tailing; see https://docs.victoriametrics.com/victorialogs/querying/#live-tailing for details", q)
+	}
+	q.Optimize()
+
+	refreshIntervalMsecs, err := httputils.GetDuration(r, "refresh_interval", 1000)
+	if err != nil {
+		httpserver.Errorf(w, r, "%s", err)
+		return
+	}
+	refreshInterval := time.Millisecond * time.Duration(refreshIntervalMsecs)
+
+	ctxWithCancel, cancel := context.WithCancel(ctx)
+	tp := newTailProcessor(cancel)
+
+	ticker := time.NewTicker(refreshInterval)
+	defer ticker.Stop()
+
+	end := time.Now().UnixNano()
+	doneCh := ctxWithCancel.Done()
+	flusher, ok := w.(http.Flusher)
+	if !ok {
+		logger.Panicf("BUG: it is expected that http.ResponseWriter (%T) supports http.Flusher interface", w)
+	}
+	for {
+		start := end - tailOffsetNsecs
+		end = time.Now().UnixNano()
+
+		qCopy := q.Clone()
+		qCopy.AddTimeFilter(start, end)
+		if err := vlstorage.RunQuery(ctxWithCancel, tenantIDs, qCopy, tp.writeBlock); err != nil {
+			httpserver.Errorf(w, r, "cannot execute tail query [%s]: %s", q, err)
+			return
+		}
+		resultRows, err := tp.getTailRows()
+		if err != nil {
+			httpserver.Errorf(w, r, "cannot get tail results for query [%q]: %s", q, err)
+			return
+		}
+		if len(resultRows) > 0 {
+			WriteJSONRows(w, resultRows)
+			flusher.Flush()
+		}
+
+		select {
+		case <-doneCh:
+			return
+		case <-ticker.C:
+		}
+	}
+}
+
+var liveTailRequests = metrics.NewCounter(`vl_live_tailing_requests`)
+
+const tailOffsetNsecs = 5e9
+
+type logRow struct {
+	timestamp int64
+	fields    []logstorage.Field
+}
+
+func sortLogRows(rows []logRow) {
+	sort.SliceStable(rows, func(i, j int) bool {
+		return rows[i].timestamp < rows[j].timestamp
+	})
+}
+
+type tailProcessor struct {
+	cancel func()
+
+	mu sync.Mutex
+
+	perStreamRows  map[string][]logRow
+	lastTimestamps map[string]int64
+
+	err error
+}
+
+func newTailProcessor(cancel func()) *tailProcessor {
+	return &tailProcessor{
+		cancel: cancel,
+
+		perStreamRows:  make(map[string][]logRow),
+		lastTimestamps: make(map[string]int64),
+	}
+}
+
+func (tp *tailProcessor) writeBlock(_ uint, timestamps []int64, columns []logstorage.BlockColumn) {
+	if len(timestamps) == 0 {
+		return
+	}
+
+	tp.mu.Lock()
+	defer tp.mu.Unlock()
+
+	if tp.err != nil {
+		return
+	}
+
+	// Make sure columns contain _time field, since it is needed for proper tail work.
+	hasTime := false
+	for _, c := range columns {
+		if c.Name == "_time" {
+			hasTime = true
+			break
+		}
+	}
+	if !hasTime {
+		tp.err = fmt.Errorf("missing _time field")
+		tp.cancel()
+		return
+	}
+
+	// Copy block rows to tp.perStreamRows
+	for i, timestamp := range timestamps {
+		streamID := ""
+		fields := make([]logstorage.Field, len(columns))
+		for j, c := range columns {
+			name := strings.Clone(c.Name)
+			value := strings.Clone(c.Values[i])
+
+			fields[j] = logstorage.Field{
+				Name:  name,
+				Value: value,
+			}
+
+			if name == "_stream_id" {
+				streamID = value
+			}
+		}
+
+		tp.perStreamRows[streamID] = append(tp.perStreamRows[streamID], logRow{
+			timestamp: timestamp,
+			fields:    fields,
+		})
+	}
+}
+
+func (tp *tailProcessor) getTailRows() ([][]logstorage.Field, error) {
+	if tp.err != nil {
+		return nil, tp.err
+	}
+
+	var resultRows []logRow
+	for streamID, rows := range tp.perStreamRows {
+		sortLogRows(rows)
+
+		lastTimestamp, ok := tp.lastTimestamps[streamID]
+		if ok {
+			// Skip already written rows
+			for len(rows) > 0 && rows[0].timestamp <= lastTimestamp {
+				rows = rows[1:]
+			}
+		}
+		if len(rows) > 0 {
+			resultRows = append(resultRows, rows...)
+			tp.lastTimestamps[streamID] = rows[len(rows)-1].timestamp
+		}
+	}
+	clear(tp.perStreamRows)
+
+	sortLogRows(resultRows)
+
+	tailRows := make([][]logstorage.Field, len(resultRows))
+	for i, row := range resultRows {
+		tailRows[i] = row.fields
+	}
+
+	return tailRows, nil
+}
+
 // ProcessQueryRequest handles /select/logsql/query request.
 //
 // See https://docs.victoriametrics.com/victorialogs/querying/#http-api
@@ -344,6 +622,7 @@ func ProcessQueryRequest(ctx context.Context, w http.ResponseWriter, r *http.Req
 
 	if err := vlstorage.RunQuery(ctx, tenantIDs, q, writeBlock); err != nil {
 		httpserver.Errorf(w, r, "cannot execute query [%s]: %s", q, err)
+		return
 	}
 }
 

app/vlselect/main.go

@@ -1,6 +1,7 @@
 package vlselect
 
 import (
+	"context"
 	"embed"
 	"flag"
 	"fmt"
@@ -13,7 +14,6 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputils"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/timerpool"
 	"github.com/VictoriaMetrics/metrics"
 )
 
@@ -23,7 +23,7 @@ var (
 		"See also -search.maxQueueDuration")
 	maxQueueDuration = flag.Duration("search.maxQueueDuration", 10*time.Second, "The maximum time the search request waits for execution when -search.maxConcurrentRequests "+
 		"limit is reached; see also -search.maxQueryDuration")
-	maxQueryDuration = flag.Duration("search.maxQueryDuration", time.Second*30, "The maximum duration for query execution")
+	maxQueryDuration = flag.Duration("search.maxQueryDuration", time.Second*30, "The maximum duration for query execution. It can be overridden on a per-query basis via 'timeout' query arg")
 )
 
 func getDefaultMaxConcurrentRequests() int {
@@ -98,47 +98,83 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 		return true
 	}
 
-	// Limit the number of concurrent queries, which can consume big amounts of CPU.
+	// Limit the number of concurrent queries, which can consume big amounts of CPU time.
 	startTime := time.Now()
 	ctx := r.Context()
-	stopCh := ctx.Done()
+	d := getMaxQueryDuration(r)
+	ctxWithTimeout, cancel := context.WithTimeout(ctx, d)
+	defer cancel()
+
+	stopCh := ctxWithTimeout.Done()
 	select {
 	case concurrencyLimitCh <- struct{}{}:
 		defer func() { <-concurrencyLimitCh }()
 	default:
 		// Sleep for a while until giving up. This should resolve short bursts in requests.
 		concurrencyLimitReached.Inc()
-		d := getMaxQueryDuration(r)
-		if d > *maxQueueDuration {
-			d = *maxQueueDuration
-		}
-		t := timerpool.Get(d)
 		select {
 		case concurrencyLimitCh <- struct{}{}:
-			timerpool.Put(t)
 			defer func() { <-concurrencyLimitCh }()
 		case <-stopCh:
-			timerpool.Put(t)
-			remoteAddr := httpserver.GetQuotedRemoteAddr(r)
-			requestURI := httpserver.GetRequestURI(r)
-			logger.Infof("client has cancelled the request after %.3f seconds: remoteAddr=%s, requestURI: %q",
-				time.Since(startTime).Seconds(), remoteAddr, requestURI)
-			return true
-		case <-t.C:
-			timerpool.Put(t)
-			concurrencyLimitTimeout.Inc()
-			err := &httpserver.ErrorWithStatusCode{
-				Err: fmt.Errorf("couldn't start executing the request in %.3f seconds, since -search.maxConcurrentRequests=%d concurrent requests "+
-					"are executed. Possible solutions: to reduce query load; to add more compute resources to the server; "+
-					"to increase -search.maxQueueDuration=%s; to increase -search.maxQueryDuration; to increase -search.maxConcurrentRequests",
-					d.Seconds(), *maxConcurrentRequests, maxQueueDuration),
-				StatusCode: http.StatusServiceUnavailable,
+			switch ctxWithTimeout.Err() {
+			case context.Canceled:
+				remoteAddr := httpserver.GetQuotedRemoteAddr(r)
+				requestURI := httpserver.GetRequestURI(r)
+				logger.Infof("client has canceled the pending request after %.3f seconds: remoteAddr=%s, requestURI: %q",
+					time.Since(startTime).Seconds(), remoteAddr, requestURI)
+			case context.DeadlineExceeded:
+				concurrencyLimitTimeout.Inc()
+				err := &httpserver.ErrorWithStatusCode{
+					Err: fmt.Errorf("couldn't start executing the request in %.3f seconds, since -search.maxConcurrentRequests=%d concurrent requests "+
+						"are executed. Possible solutions: to reduce query load; to add more compute resources to the server; "+
+						"to increase -search.maxQueueDuration=%s; to increase -search.maxQueryDuration=%s; to increase -search.maxConcurrentRequests; "+
+						"to pass bigger value to 'timeout' query arg",
+						d.Seconds(), *maxConcurrentRequests, maxQueueDuration, maxQueryDuration),
+					StatusCode: http.StatusServiceUnavailable,
+				}
+				httpserver.Errorf(w, r, "%s", err)
 			}
-			httpserver.Errorf(w, r, "%s", err)
 			return true
 		}
 	}
 
+	if path == "/select/logsql/tail" {
+		logsqlTailRequests.Inc()
+		// Process live tailing request without timeout (e.g. use ctx instead of ctxWithTimeout),
+		// since it is OK to run live tailing requests for very long time.
+		logsql.ProcessLiveTailRequest(ctx, w, r)
+		return true
+	}
+
+	ok := processSelectRequest(ctxWithTimeout, w, r, path)
+	if !ok {
+		return false
+	}
+
+	err := ctxWithTimeout.Err()
+	switch err {
+	case nil:
+		// nothing to do
+	case context.Canceled:
+		remoteAddr := httpserver.GetQuotedRemoteAddr(r)
+		requestURI := httpserver.GetRequestURI(r)
+		logger.Infof("client has canceled the request after %.3f seconds: remoteAddr=%s, requestURI: %q",
+			time.Since(startTime).Seconds(), remoteAddr, requestURI)
+	case context.DeadlineExceeded:
+		err = &httpserver.ErrorWithStatusCode{
+			Err: fmt.Errorf("the request couldn't be executed in %.3f seconds; possible solutions: "+
+				"to increase -search.maxQueryDuration=%s; to pass bigger value to 'timeout' query arg", d.Seconds(), maxQueryDuration),
+			StatusCode: http.StatusServiceUnavailable,
+		}
+		httpserver.Errorf(w, r, "%s", err)
+	default:
+		httpserver.Errorf(w, r, "unexpected error: %s", err)
+	}
+
+	return true
+}
+
+func processSelectRequest(ctx context.Context, w http.ResponseWriter, r *http.Request, path string) bool {
 	httpserver.EnableCORS(w, r)
 	switch path {
 	case "/select/logsql/field_names":
@@ -165,6 +201,10 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
 		logsqlStreamFieldValuesRequests.Inc()
 		logsql.ProcessStreamFieldValuesRequest(ctx, w, r)
 		return true
+	case "/select/logsql/stream_ids":
+		logsqlStreamIDsRequests.Inc()
+		logsql.ProcessStreamIDsRequest(ctx, w, r)
+		return true
 	case "/select/logsql/streams":
 		logsqlStreamsRequests.Inc()
 		logsql.ProcessStreamsRequest(ctx, w, r)
@@ -194,5 +234,7 @@ var (
 	logsqlQueryRequests             = metrics.NewCounter(`vl_http_requests_total{path="/select/logsql/query"}`)
 	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"}`)
 	logsqlStreamsRequests           = metrics.NewCounter(`vl_http_requests_total{path="/select/logsql/streams"}`)
+	logsqlTailRequests              = metrics.NewCounter(`vl_http_requests_total{path="/select/logsql/tail"}`)
 )

app/vlselect/vmui/asset-manifest.json

@@ -1,13 +1,13 @@
 {
   "files": {
-    "main.css": "./static/css/main.2fa7c03f.css",
-    "main.js": "./static/js/main.68f1bd69.js",
+    "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.cb83d071da309a358bc0.md",
+    "static/media/MetricsQL.md": "./static/media/MetricsQL.aaabf95f2c9bf356bde4.md",
     "index.html": "./index.html"
   },
   "entrypoints": [
-    "static/css/main.2fa7c03f.css",
-    "static/js/main.68f1bd69.js"
+    "static/css/main.1041c3d4.css",
+    "static/js/main.62a82f4a.js"
   ]
 }

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.68f1bd69.js"></script><link href="./static/css/main.2fa7c03f.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.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>

app/vlselect/vmui/static/js/main.62a82f4a.js.LICENSE.txt

@@ -5,7 +5,7 @@
 */
 
 /**
- * @remix-run/router v1.15.1
+ * @remix-run/router v1.17.0
  *
  * Copyright (c) Remix Software Inc.
  *
@@ -16,7 +16,7 @@
  */
 
 /**
- * React Router DOM v6.22.1
+ * React Router DOM v6.24.0
  *
  * Copyright (c) Remix Software Inc.
  *
@@ -27,7 +27,7 @@
  */
 
 /**
- * React Router v6.22.1
+ * React Router v6.24.0
  *
  * Copyright (c) Remix Software Inc.
  *

app/vlselect/vmui/static/media/MetricsQL.aaabf95f2c9bf356bde4.md

@@ -107,7 +107,7 @@ 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 value MetricsQL expression.
+* Metric names and label names may contain any unicode letter. For example `температура{город="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:

app/vlstorage/main.go

@@ -20,10 +20,12 @@ import (
 var (
 	retentionPeriod = flagutil.NewDuration("retentionPeriod", "7d", "Log entries with timestamps older than now-retentionPeriod are automatically deleted; "+
 		"log entries with timestamps outside the retention are also rejected during data ingestion; the minimum supported retention is 1d (one day); "+
-		"see https://docs.victoriametrics.com/victorialogs/#retention")
+		"see https://docs.victoriametrics.com/victorialogs/#retention ; see also -retention.maxDiskSpaceUsageBytes")
+	maxDiskSpaceUsageBytes = flagutil.NewBytes("retention.maxDiskSpaceUsageBytes", 0, "The maximum disk space usage at -storageDataPath before older per-day "+
+		"partitions are automatically dropped; see https://docs.victoriametrics.com/victorialogs/#retention-by-disk-space-usage ; see also -retentionPeriod")
 	futureRetention = flagutil.NewDuration("futureRetention", "2d", "Log entries with timestamps bigger than now+futureRetention are rejected during data ingestion; "+
 		"see https://docs.victoriametrics.com/victorialogs/#retention")
-	storageDataPath = flag.String("storageDataPath", "victoria-logs-data", "Path to directory with the VictoriaLogs data; "+
+	storageDataPath = flag.String("storageDataPath", "victoria-logs-data", "Path to directory where to store VictoriaLogs data; "+
 		"see https://docs.victoriametrics.com/victorialogs/#storage")
 	inmemoryDataFlushInterval = flag.Duration("inmemoryDataFlushInterval", 5*time.Second, "The interval for guaranteed saving of in-memory data to disk. "+
 		"The saved data survives unclean shutdowns such as OOM crash, hardware reset, SIGKILL, etc. "+
@@ -49,12 +51,13 @@ func Init() {
 		logger.Fatalf("-retentionPeriod cannot be smaller than a day; got %s", retentionPeriod)
 	}
 	cfg := &logstorage.StorageConfig{
-		Retention:             retentionPeriod.Duration(),
-		FlushInterval:         *inmemoryDataFlushInterval,
-		FutureRetention:       futureRetention.Duration(),
-		LogNewStreams:         *logNewStreams,
-		LogIngestedRows:       *logIngestedRows,
-		MinFreeDiskSpaceBytes: minFreeDiskSpaceBytes.N,
+		Retention:              retentionPeriod.Duration(),
+		MaxDiskSpaceUsageBytes: maxDiskSpaceUsageBytes.N,
+		FlushInterval:          *inmemoryDataFlushInterval,
+		FutureRetention:        futureRetention.Duration(),
+		LogNewStreams:          *logNewStreams,
+		LogIngestedRows:        *logIngestedRows,
+		MinFreeDiskSpaceBytes:  minFreeDiskSpaceBytes.N,
 	}
 	logger.Infof("opening storage at -storageDataPath=%s", *storageDataPath)
 	startTime := time.Now()
@@ -77,7 +80,7 @@ func Init() {
 
 // Stop stops vlstorage.
 func Stop() {
-	metrics.UnregisterSet(storageMetrics)
+	metrics.UnregisterSet(storageMetrics, true)
 	storageMetrics = nil
 
 	strg.MustClose()
@@ -142,6 +145,13 @@ func GetStreams(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstor
 	return strg.GetStreams(ctx, tenantIDs, q, limit)
 }
 
+// GetStreamIDs executes q and returns streamIDs seen in query results.
+//
+// If limit > 0, then up to limit unique streamIDs are returned.
+func GetStreamIDs(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, limit uint64) ([]logstorage.ValueWithHits, error) {
+	return strg.GetStreamIDs(ctx, tenantIDs, q, limit)
+}
+
 func writeStorageMetrics(w io.Writer, strg *logstorage.Storage) {
 	var ss logstorage.StorageStats
 	strg.UpdateStats(&ss)

app/vmagent/main.go

@@ -24,7 +24,6 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/prometheusimport"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/promremotewrite"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/remotewrite"
-	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/statsd"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/vmimport"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/auth"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/buildinfo"
@@ -37,7 +36,6 @@ import (
 	influxserver "github.com/VictoriaMetrics/VictoriaMetrics/lib/ingestserver/influx"
 	opentsdbserver "github.com/VictoriaMetrics/VictoriaMetrics/lib/ingestserver/opentsdb"
 	opentsdbhttpserver "github.com/VictoriaMetrics/VictoriaMetrics/lib/ingestserver/opentsdbhttp"
-	statsdserver "github.com/VictoriaMetrics/VictoriaMetrics/lib/ingestserver/statsd"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/procutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promscrape"
@@ -63,12 +61,6 @@ var (
 		"See also -graphiteListenAddr.useProxyProtocol")
 	graphiteUseProxyProtocol = flag.Bool("graphiteListenAddr.useProxyProtocol", false, "Whether to use proxy protocol for connections accepted at -graphiteListenAddr . "+
 		"See https://www.haproxy.org/download/1.8/doc/proxy-protocol.txt")
-	statsdListenAddr = flag.String("statsdListenAddr", "", "TCP and UDP address to listen for Statsd plaintext data. Usually :8125 must be set. Doesn't work if empty. "+
-		"See also -statsdListenAddr.useProxyProtocol")
-	statsdUseProxyProtocol = flag.Bool("statsdListenAddr.useProxyProtocol", false, "Whether to use proxy protocol for connections accepted at -statsdListenAddr . "+
-		"See https://www.haproxy.org/download/1.8/doc/proxy-protocol.txt")
-	statsdDisableAggregationEnforcemenet = flag.Bool(`statsd.disableAggregationEnforcement`, false, "Whether to disable streaming aggregation requirement check. "+
-		"It's recommended to run statsdServer with pre-configured streaming aggregation to decrease load at database.")
 	opentsdbListenAddr = flag.String("opentsdbListenAddr", "", "TCP and UDP address to listen for OpenTSDB metrics. "+
 		"Telnet put messages and HTTP /api/put messages are simultaneously served on TCP port. "+
 		"Usually :4242 must be set. Doesn't work if empty. See also -opentsdbListenAddr.useProxyProtocol")
@@ -78,8 +70,8 @@ var (
 		"See also -opentsdbHTTPListenAddr.useProxyProtocol")
 	opentsdbHTTPUseProxyProtocol = flag.Bool("opentsdbHTTPListenAddr.useProxyProtocol", false, "Whether to use proxy protocol for connections accepted "+
 		"at -opentsdbHTTPListenAddr . See https://www.haproxy.org/download/1.8/doc/proxy-protocol.txt")
-	configAuthKey = flagutil.NewPassword("configAuthKey", "Authorization key for accessing /config page. It must be passed via authKey query arg. It overrides httpAuth.* settings.")
-	reloadAuthKey = flagutil.NewPassword("reloadAuthKey", "Auth key for /-/reload http endpoint. It must be passed via authKey query arg. It overrides httpAuth.* settings.")
+	configAuthKey = flagutil.NewPassword("configAuthKey", "Authorization key for accessing /config page. It must be passed via authKey query arg. It overrides -httpAuth.*")
+	reloadAuthKey = flagutil.NewPassword("reloadAuthKey", "Auth key for /-/reload http endpoint. It must be passed via authKey query arg. It overrides -httpAuth.*")
 	dryRun        = flag.Bool("dryRun", false, "Whether to check config files without running vmagent. The following files are checked: "+
 		"-promscrape.config, -remoteWrite.relabelConfig, -remoteWrite.urlRelabelConfig, -remoteWrite.streamAggr.config . "+
 		"Unknown config entries aren't allowed in -promscrape.config by default. This can be changed by passing -promscrape.config.strictParse=false command-line flag")
@@ -88,7 +80,6 @@ var (
 var (
 	influxServer       *influxserver.Server
 	graphiteServer     *graphiteserver.Server
-	statsdServer       *statsdserver.Server
 	opentsdbServer     *opentsdbserver.Server
 	opentsdbhttpServer *opentsdbhttpserver.Server
 )
@@ -123,7 +114,7 @@ func main() {
 			logger.Fatalf("error when checking relabel configs: %s", err)
 		}
 		if err := remotewrite.CheckStreamAggrConfigs(); err != nil {
-			logger.Fatalf("error when checking -remoteWrite.streamAggr.config: %s", err)
+			logger.Fatalf("error when checking -streamAggr.config and -remoteWrite.streamAggr.config: %s", err)
 		}
 		logger.Infof("all the configs are ok; exiting with 0 status code")
 		return
@@ -146,12 +137,6 @@ func main() {
 	if len(*graphiteListenAddr) > 0 {
 		graphiteServer = graphiteserver.MustStart(*graphiteListenAddr, *graphiteUseProxyProtocol, graphite.InsertHandler)
 	}
-	if len(*statsdListenAddr) > 0 {
-		if !remotewrite.HasAnyStreamAggrConfigured() && !*statsdDisableAggregationEnforcemenet {
-			logger.Fatalf("streaming aggregation must be configured with enabled statsd server. It's recommended  to aggregate metrics received at statsd listener. This check could be disabled with flag -statsd.disableAggregationEnforcement")
-		}
-		statsdServer = statsdserver.MustStart(*statsdListenAddr, *statsdUseProxyProtocol, statsd.InsertHandler)
-	}
 	if len(*opentsdbListenAddr) > 0 {
 		httpInsertHandler := getOpenTSDBHTTPInsertHandler()
 		opentsdbServer = opentsdbserver.MustStart(*opentsdbListenAddr, *opentsdbUseProxyProtocol, opentsdb.InsertHandler, httpInsertHandler)
@@ -187,9 +172,6 @@ func main() {
 	if len(*graphiteListenAddr) > 0 {
 		graphiteServer.MustStop()
 	}
-	if len(*statsdListenAddr) > 0 {
-		statsdServer.MustStop()
-	}
 	if len(*opentsdbListenAddr) > 0 {
 		opentsdbServer.MustStop()
 	}
@@ -452,7 +434,7 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 		}
 		return true
 	case "/prometheus/config", "/config":
-		if !httpserver.CheckAuthFlag(w, r, configAuthKey.Get(), "configAuthKey") {
+		if !httpserver.CheckAuthFlag(w, r, configAuthKey) {
 			return true
 		}
 		promscrapeConfigRequests.Inc()
@@ -461,7 +443,7 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 		return true
 	case "/prometheus/api/v1/status/config", "/api/v1/status/config":
 		// See https://prometheus.io/docs/prometheus/latest/querying/api/#config
-		if !httpserver.CheckAuthFlag(w, r, configAuthKey.Get(), "configAuthKey") {
+		if !httpserver.CheckAuthFlag(w, r, configAuthKey) {
 			return true
 		}
 		promscrapeStatusConfigRequests.Inc()
@@ -471,7 +453,7 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 		fmt.Fprintf(w, `{"status":"success","data":{"yaml":%q}}`, bb.B)
 		return true
 	case "/prometheus/-/reload", "/-/reload":
-		if !httpserver.CheckAuthFlag(w, r, reloadAuthKey.Get(), "reloadAuthKey") {
+		if !httpserver.CheckAuthFlag(w, r, reloadAuthKey) {
 			return true
 		}
 		promscrapeConfigReloadRequests.Inc()

app/vmagent/multiarch/Dockerfile

@@ -1,7 +1,7 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
 ARG certs_image
 ARG root_image
-FROM $certs_image as certs
+FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 
 FROM $root_image

app/vmagent/prometheusimport/request_handler_test.go

@@ -26,11 +26,11 @@ func TestInsertHandler(t *testing.T) {
 	req := httptest.NewRequest(http.MethodPost, "/insert/0/api/v1/import/prometheus", bytes.NewBufferString(`{"foo":"bar"}
 go_memstats_alloc_bytes_total 1`))
 	if err := InsertHandler(nil, req); err != nil {
-		t.Errorf("unxepected error %s", err)
+		t.Fatalf("unxepected error %s", err)
 	}
 	expectedMsg := "cannot unmarshal Prometheus line"
 	if !strings.Contains(testOutput.String(), expectedMsg) {
-		t.Errorf("output %q should contain %q", testOutput.String(), expectedMsg)
+		t.Fatalf("output %q should contain %q", testOutput.String(), expectedMsg)
 	}
 }
 

app/vmagent/remotewrite/client.go

@@ -15,8 +15,8 @@ import (
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/awsapi"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputils"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/netutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/persistentqueue"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promauth"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
@@ -120,7 +120,7 @@ func newHTTPClient(argIdx int, remoteWriteURL, sanitizedURL string, fq *persiste
 		logger.Fatalf("cannot initialize AWS Config for -remoteWrite.url=%q: %s", remoteWriteURL, err)
 	}
 	tr := &http.Transport{
-		DialContext:         httputils.GetStatDialFunc("vmagent_remotewrite"),
+		DialContext:         netutil.NewStatDialFunc("vmagent_remotewrite"),
 		TLSHandshakeTimeout: tlsHandshakeTimeout.GetOptionalArg(argIdx),
 		MaxConnsPerHost:     2 * concurrency,
 		MaxIdleConnsPerHost: 2 * concurrency,

app/vmagent/remotewrite/pendingseries.go

@@ -109,10 +109,9 @@ type writeRequest struct {
 
 	wr prompbmarshal.WriteRequest
 
-	tss       []prompbmarshal.TimeSeries
-	labels    []prompbmarshal.Label
-	samples   []prompbmarshal.Sample
-	exemplars []prompbmarshal.Exemplar
+	tss     []prompbmarshal.TimeSeries
+	labels  []prompbmarshal.Label
+	samples []prompbmarshal.Sample
 
 	// buf holds labels data
 	buf []byte
@@ -130,7 +129,6 @@ func (wr *writeRequest) reset() {
 	wr.labels = wr.labels[:0]
 
 	wr.samples = wr.samples[:0]
-	wr.exemplars = wr.exemplars[:0]
 	wr.buf = wr.buf[:0]
 }
 
@@ -202,7 +200,6 @@ func (wr *writeRequest) copyTimeSeries(dst, src *prompbmarshal.TimeSeries) {
 	labelsDst := wr.labels
 	labelsLen := len(wr.labels)
 	samplesDst := wr.samples
-	exemplarsDst := wr.exemplars
 	buf := wr.buf
 	for i := range src.Labels {
 		labelsDst = append(labelsDst, prompbmarshal.Label{})
@@ -219,12 +216,8 @@ func (wr *writeRequest) copyTimeSeries(dst, src *prompbmarshal.TimeSeries) {
 	samplesDst = append(samplesDst, src.Samples...)
 	dst.Samples = samplesDst[len(samplesDst)-len(src.Samples):]
 
-	exemplarsDst = append(exemplarsDst, src.Exemplars...)
-	dst.Exemplars = exemplarsDst[len(exemplarsDst)-len(src.Exemplars):]
-
 	wr.samples = samplesDst
 	wr.labels = labelsDst
-	wr.exemplars = exemplarsDst
 	wr.buf = buf
 }
 
@@ -236,6 +229,7 @@ func tryPushWriteRequest(wr *prompbmarshal.WriteRequest, tryPushBlock func(block
 		// Nothing to push
 		return true
 	}
+
 	marshalConcurrencyCh <- struct{}{}
 
 	bb := writeRequestBufPool.Get()
@@ -272,8 +266,6 @@ func tryPushWriteRequest(wr *prompbmarshal.WriteRequest, tryPushBlock func(block
 	if len(wr.Timeseries) == 1 {
 		// A single time series left. Recursively split its samples into smaller parts if possible.
 		samples := wr.Timeseries[0].Samples
-		exemplars := wr.Timeseries[0].Exemplars
-
 		if len(samples) == 1 {
 			logger.Warnf("dropping a sample for metric with too long labels exceeding -remoteWrite.maxBlockSize=%d bytes", maxUnpackedBlockSize.N)
 			return true
@@ -285,16 +277,11 @@ func tryPushWriteRequest(wr *prompbmarshal.WriteRequest, tryPushBlock func(block
 			return false
 		}
 		wr.Timeseries[0].Samples = samples[n:]
-		// We do not want to send exemplars twice
-		wr.Timeseries[0].Exemplars = nil
-
 		if !tryPushWriteRequest(wr, tryPushBlock, isVMRemoteWrite) {
 			wr.Timeseries[0].Samples = samples
-			wr.Timeseries[0].Exemplars = exemplars
 			return false
 		}
 		wr.Timeseries[0].Samples = samples
-		wr.Timeseries[0].Exemplars = exemplars
 		return true
 	}
 	timeseries := wr.Timeseries

app/vmagent/remotewrite/pendingseries_test.go

@@ -10,8 +10,8 @@ import (
 
 func TestPushWriteRequest(t *testing.T) {
 	rowsCounts := []int{1, 10, 100, 1e3, 1e4}
-	expectedBlockLensProm := []int{248, 1952, 17433, 180381, 1861994}
-	expectedBlockLensVM := []int{170, 575, 4748, 44936, 367096}
+	expectedBlockLensProm := []int{216, 1848, 16424, 169882, 1757876}
+	expectedBlockLensVM := []int{138, 492, 3927, 34995, 288476}
 	for i, rowsCount := range rowsCounts {
 		expectedBlockLenProm := expectedBlockLensProm[i]
 		expectedBlockLenVM := expectedBlockLensVM[i]
@@ -59,20 +59,6 @@ func newTestWriteRequest(seriesCount, labelsCount int) *prompbmarshal.WriteReque
 				Value: fmt.Sprintf("value_%d_%d", i, j),
 			})
 		}
-		exemplar := prompbmarshal.Exemplar{
-			Labels: []prompbmarshal.Label{
-				{
-					Name:  "trace_id",
-					Value: "123456",
-				},
-				{
-					Name:  "log_id",
-					Value: "987654",
-				},
-			},
-			Value:     float64(i),
-			Timestamp: 1000 * int64(i),
-		}
 		wr.Timeseries = append(wr.Timeseries, prompbmarshal.TimeSeries{
 			Labels: labels,
 			Samples: []prompbmarshal.Sample{
@@ -81,10 +67,6 @@ func newTestWriteRequest(seriesCount, labelsCount int) *prompbmarshal.WriteReque
 					Timestamp: 1000 * int64(i),
 				},
 			},
-
-			Exemplars: []prompbmarshal.Exemplar{
-				exemplar,
-			},
 		})
 	}
 	return &wr

app/vmagent/remotewrite/relabel.go

@@ -181,7 +181,7 @@ func (rctx *relabelCtx) reset() {
 }
 
 var relabelCtxPool = &sync.Pool{
-	New: func() interface{} {
+	New: func() any {
 		return &relabelCtx{}
 	},
 }

app/vmagent/remotewrite/remotewrite.go

@@ -6,6 +6,7 @@ import (
 	"net/http"
 	"net/url"
 	"path/filepath"
+	"slices"
 	"strconv"
 	"sync"
 	"sync/atomic"
@@ -88,15 +89,15 @@ var (
 		"By default there are no limits on samples ingestion rate. See also -remoteWrite.rateLimit")
 
 	disableOnDiskQueue = flagutil.NewArrayBool("remoteWrite.disableOnDiskQueue", "Whether to disable storing pending data to -remoteWrite.tmpDataPath "+
-		"when the configured remote storage systems cannot keep up with the data ingestion rate. See https://docs.victoriametrics.com/vmagent#disabling-on-disk-persistence ."+
-		"See also -remoteWrite.dropSamplesOnOverload")
-	dropSamplesOnOverload = flagutil.NewArrayBool("remoteWrite.dropSamplesOnOverload", "Whether to drop samples when -remoteWrite.disableOnDiskQueue is set and if the samples "+
-		"cannot be pushed into the configured remote storage systems in a timely manner. See https://docs.victoriametrics.com/vmagent#disabling-on-disk-persistence")
+		"when the remote storage system at the corresponding -remoteWrite.url cannot keep up with the data ingestion rate. "+
+		"See https://docs.victoriametrics.com/vmagent#disabling-on-disk-persistence . See also -remoteWrite.dropSamplesOnOverload")
+	dropSamplesOnOverload = flag.Bool("remoteWrite.dropSamplesOnOverload", false, "Whether to drop samples when -remoteWrite.disableOnDiskQueue is set and if the samples "+
+		"cannot be pushed into the configured -remoteWrite.url systems in a timely manner. See https://docs.victoriametrics.com/vmagent#disabling-on-disk-persistence")
 )
 
 var (
-	// rwctxs contains statically populated entries when -remoteWrite.url is specified.
-	rwctxs []*remoteWriteCtx
+	// rwctxsGlobal contains statically populated entries when -remoteWrite.url is specified.
+	rwctxsGlobal []*remoteWriteCtx
 
 	// Data without tenant id is written to defaultAuthToken if -enableMultitenantHandlers is specified.
 	defaultAuthToken = &auth.Token{}
@@ -109,8 +110,11 @@ var (
 		StatusCode: http.StatusTooManyRequests,
 	}
 
-	// disableOnDiskQueueAll is set to true if all remoteWrite.urls were configured to disable persistent queue via disableOnDiskQueue
-	disableOnDiskQueueAll bool
+	// disableOnDiskQueueAny is set to true if at least a single -remoteWrite.url is configured with -remoteWrite.disableOnDiskQueue
+	disableOnDiskQueueAny bool
+
+	// dropSamplesOnFailureGlobal is set to true if -remoteWrite.dropSamplesOnOverload is set or if multiple -remoteWrite.disableOnDiskQueue options are set.
+	dropSamplesOnFailureGlobal bool
 )
 
 // MultitenancyEnabled returns true if -enableMultitenantHandlers is specified.
@@ -203,28 +207,18 @@ func Init() {
 	relabelConfigSuccess.Set(1)
 	relabelConfigTimestamp.Set(fasttime.UnixTimestamp())
 
-	sasFile, sasOpts := getStreamAggrOpts(-1)
-	if sasFile != "" {
-		sas, err := newStreamAggrConfig(-1, pushToRemoteStoragesDropFailed)
-		if err != nil {
-			logger.Fatalf("cannot initialize stream aggregators from -streamAggr.config=%q: %s", sasFile, err)
-		}
-		sasGlobal.Store(sas)
-	} else if sasOpts.DedupInterval > 0 {
-		deduplicatorGlobal = streamaggr.NewDeduplicator(pushToRemoteStoragesDropFailed, sasOpts.DedupInterval, sasOpts.DropInputLabels)
-	}
+	initStreamAggrConfigGlobal()
 
-	if len(*remoteWriteURLs) > 0 {
-		rwctxs = newRemoteWriteCtxs(nil, *remoteWriteURLs)
-	}
+	rwctxsGlobal = newRemoteWriteCtxs(nil, *remoteWriteURLs)
 
-	disableOnDiskQueueAll = true
-	for _, v := range *disableOnDiskQueue {
-		if !v {
-			disableOnDiskQueueAll = false
-			break
-		}
-	}
+	disableOnDiskQueues := []bool(*disableOnDiskQueue)
+	disableOnDiskQueueAny = slices.Contains(disableOnDiskQueues, true)
+
+	// Samples must be dropped if multiple -remoteWrite.disableOnDiskQueue options are configured and at least a single is set to true.
+	// In this case it is impossible to prevent from sending many duplicates of samples passed to TryPush() to all the configured -remoteWrite.url
+	// if these samples couldn't be sent to the -remoteWrite.url with the disabled persistent queue. So it is better sending samples
+	// to the remaining -remoteWrite.url and dropping them on the blocked queue.
+	dropSamplesOnFailureGlobal = *dropSamplesOnOverload || disableOnDiskQueueAny && len(disableOnDiskQueues) > 1
 
 	dropDanglingQueues()
 
@@ -234,9 +228,9 @@ func Init() {
 		defer configReloaderWG.Done()
 		for {
 			select {
-			case <-sighupCh:
 			case <-configReloaderStopCh:
 				return
+			case <-sighupCh:
 			}
 			reloadRelabelConfigs()
 			reloadStreamAggrConfigs()
@@ -255,8 +249,8 @@ func dropDanglingQueues() {
 	// In case if there were many persistent queues with identical *remoteWriteURLs
 	// the queue with the last index will be dropped.
 	// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/6140
-	existingQueues := make(map[string]struct{}, len(rwctxs))
-	for _, rwctx := range rwctxs {
+	existingQueues := make(map[string]struct{}, len(rwctxsGlobal))
+	for _, rwctx := range rwctxsGlobal {
 		existingQueues[rwctx.fq.Dirname()] = struct{}{}
 	}
 
@@ -273,7 +267,7 @@ func dropDanglingQueues() {
 		}
 	}
 	if removed > 0 {
-		logger.Infof("removed %d dangling queues from %q, active queues: %d", removed, *tmpDataPath, len(rwctxs))
+		logger.Infof("removed %d dangling queues from %q, active queues: %d", removed, *tmpDataPath, len(rwctxsGlobal))
 	}
 }
 
@@ -382,10 +376,10 @@ func Stop() {
 		deduplicatorGlobal = nil
 	}
 
-	for _, rwctx := range rwctxs {
+	for _, rwctx := range rwctxsGlobal {
 		rwctx.MustStop()
 	}
-	rwctxs = nil
+	rwctxsGlobal = nil
 
 	if sl := hourlySeriesLimiter; sl != nil {
 		sl.MustStop()
@@ -397,6 +391,8 @@ func Stop() {
 
 // PushDropSamplesOnFailure pushes wr to the configured remote storage systems set via -remoteWrite.url
 //
+// PushDropSamplesOnFailure drops wr samples if they cannot be sent to -remoteWrite.url by any reason.
+//
 // PushDropSamplesOnFailure can modify wr contents.
 func PushDropSamplesOnFailure(at *auth.Token, wr *prompbmarshal.WriteRequest) {
 	_ = tryPush(at, wr, true)
@@ -409,7 +405,7 @@ func PushDropSamplesOnFailure(at *auth.Token, wr *prompbmarshal.WriteRequest) {
 //
 // The caller must return ErrQueueFullHTTPRetry to the client, which sends wr, if TryPush returns false.
 func TryPush(at *auth.Token, wr *prompbmarshal.WriteRequest) bool {
-	return tryPush(at, wr, false)
+	return tryPush(at, wr, dropSamplesOnFailureGlobal)
 }
 
 func tryPush(at *auth.Token, wr *prompbmarshal.WriteRequest, forceDropSamplesOnFailure bool) bool {
@@ -426,24 +422,20 @@ func tryPush(at *auth.Token, wr *prompbmarshal.WriteRequest, forceDropSamplesOnF
 		tenantRctx = getRelabelCtx()
 		defer putRelabelCtx(tenantRctx)
 	}
-	rowsCount := getRowsCount(tss)
 
 	// Quick check whether writes to configured remote storage systems are blocked.
 	// This allows saving CPU time spent on relabeling and block compression
 	// if some of remote storage systems cannot keep up with the data ingestion rate.
-	// this shortcut is only applicable if all remote writes have disableOnDiskQueue = true
-	if disableOnDiskQueueAll {
-		for _, rwctx := range rwctxs {
-			if rwctx.fq.IsWriteBlocked() {
-				rwctx.pushFailures.Inc()
-				if forceDropSamplesOnFailure || rwctx.dropSamplesOnOverload {
-					// Just drop samples
-					rwctx.rowsDroppedOnPushFailure.Add(rowsCount)
-					continue
-				}
-				return false
-			}
-		}
+	rwctxs, ok := getEligibleRemoteWriteCtxs(tss, forceDropSamplesOnFailure)
+	if !ok {
+		// At least a single remote write queue is blocked and dropSamplesOnFailure isn't set.
+		// Return false to the caller, so it could re-send samples again.
+		return false
+	}
+	if len(rwctxs) == 0 {
+		// All the remote write queues are skipped because they are blocked and dropSamplesOnFailure is set to true.
+		// Return true to the caller, so it doesn't re-send the samples again.
+		return true
 	}
 
 	var rctx *relabelCtx
@@ -453,6 +445,7 @@ func tryPush(at *auth.Token, wr *prompbmarshal.WriteRequest, forceDropSamplesOnF
 		rctx = getRelabelCtx()
 		defer putRelabelCtx(rctx)
 	}
+	rowsCount := getRowsCount(tss)
 	globalRowsPushedBeforeRelabel.Add(rowsCount)
 	maxSamplesPerBlock := *maxRowsPerBlock
 	// Allow up to 10x of labels per each block on average.
@@ -505,20 +498,46 @@ func tryPush(at *auth.Token, wr *prompbmarshal.WriteRequest, forceDropSamplesOnF
 			deduplicatorGlobal.Push(tssBlock)
 			tssBlock = tssBlock[:0]
 		}
-		if !tryPushBlockToRemoteStorages(tssBlock, forceDropSamplesOnFailure) {
+		if !tryPushBlockToRemoteStorages(rwctxs, tssBlock, forceDropSamplesOnFailure) {
 			return false
 		}
 	}
 	return true
 }
 
-func pushToRemoteStoragesDropFailed(tss []prompbmarshal.TimeSeries) {
-	if tryPushBlockToRemoteStorages(tss, true) {
+func getEligibleRemoteWriteCtxs(tss []prompbmarshal.TimeSeries, forceDropSamplesOnFailure bool) ([]*remoteWriteCtx, bool) {
+	if !disableOnDiskQueueAny {
+		return rwctxsGlobal, true
+	}
+
+	// This code is applicable if at least a single remote storage has -disableOnDiskQueue
+	rwctxs := make([]*remoteWriteCtx, 0, len(rwctxsGlobal))
+	for _, rwctx := range rwctxsGlobal {
+		if !rwctx.fq.IsWriteBlocked() {
+			rwctxs = append(rwctxs, rwctx)
+		} else {
+			rwctx.pushFailures.Inc()
+			if !forceDropSamplesOnFailure {
+				return nil, false
+			}
+			rowsCount := getRowsCount(tss)
+			rwctx.rowsDroppedOnPushFailure.Add(rowsCount)
+		}
+	}
+	return rwctxs, true
+}
+
+func pushToRemoteStoragesTrackDropped(tss []prompbmarshal.TimeSeries) {
+	rwctxs, _ := getEligibleRemoteWriteCtxs(tss, true)
+	if len(rwctxs) == 0 {
 		return
 	}
+	if !tryPushBlockToRemoteStorages(rwctxs, tss, true) {
+		logger.Panicf("BUG: tryPushBlockToRemoteStorages() must return true when forceDropSamplesOnFailure=true")
+	}
 }
 
-func tryPushBlockToRemoteStorages(tssBlock []prompbmarshal.TimeSeries, forceDropSamplesOnFailure bool) bool {
+func tryPushBlockToRemoteStorages(rwctxs []*remoteWriteCtx, tssBlock []prompbmarshal.TimeSeries, forceDropSamplesOnFailure bool) bool {
 	if len(tssBlock) == 0 {
 		// Nothing to push
 		return true
@@ -537,7 +556,7 @@ func tryPushBlockToRemoteStorages(tssBlock []prompbmarshal.TimeSeries, forceDrop
 		if replicas <= 0 {
 			replicas = 1
 		}
-		return tryShardingBlockAmongRemoteStorages(tssBlock, replicas, forceDropSamplesOnFailure)
+		return tryShardingBlockAmongRemoteStorages(rwctxs, tssBlock, replicas, forceDropSamplesOnFailure)
 	}
 
 	// Replicate tssBlock samples among rwctxs.
@@ -558,7 +577,7 @@ func tryPushBlockToRemoteStorages(tssBlock []prompbmarshal.TimeSeries, forceDrop
 	return !anyPushFailed.Load()
 }
 
-func tryShardingBlockAmongRemoteStorages(tssBlock []prompbmarshal.TimeSeries, replicas int, forceDropSamplesOnFailure bool) bool {
+func tryShardingBlockAmongRemoteStorages(rwctxs []*remoteWriteCtx, tssBlock []prompbmarshal.TimeSeries, replicas int, forceDropSamplesOnFailure bool) bool {
 	x := getTSSShards(len(rwctxs))
 	defer putTSSShards(x)
 
@@ -745,10 +764,8 @@ type remoteWriteCtx struct {
 	sas          atomic.Pointer[streamaggr.Aggregators]
 	deduplicator *streamaggr.Deduplicator
 
-	streamAggrKeepInput   bool
-	streamAggrDropInput   bool
-	disableOnDiskQueue    bool
-	dropSamplesOnOverload bool
+	streamAggrKeepInput bool
+	streamAggrDropInput bool
 
 	pss        []*pendingSeries
 	pssNextIdx atomic.Uint64
@@ -773,6 +790,7 @@ func newRemoteWriteCtx(argIdx int, remoteWriteURL *url.URL, maxInmemoryBlocks in
 		logger.Warnf("rounding the -remoteWrite.maxDiskUsagePerURL=%d to the minimum supported value: %d", maxPendingBytes, persistentqueue.DefaultChunkFileSize)
 		maxPendingBytes = persistentqueue.DefaultChunkFileSize
 	}
+
 	isPQDisabled := disableOnDiskQueue.GetOptionalArg(argIdx)
 	fq := persistentqueue.MustOpenFastQueue(queuePath, sanitizedURL, maxInmemoryBlocks, maxPendingBytes, isPQDisabled)
 	_ = metrics.GetOrCreateGauge(fmt.Sprintf(`vmagent_remotewrite_pending_data_bytes{path=%q, url=%q}`, queuePath, sanitizedURL), func() float64 {
@@ -817,38 +835,20 @@ func newRemoteWriteCtx(argIdx int, remoteWriteURL *url.URL, maxInmemoryBlocks in
 		c:   c,
 		pss: pss,
 
-		dropSamplesOnOverload: dropSamplesOnOverload.GetOptionalArg(argIdx),
-		disableOnDiskQueue:    isPQDisabled,
+		rowsPushedAfterRelabel: metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_rows_pushed_after_relabel_total{path=%q,url=%q}`, queuePath, sanitizedURL)),
+		rowsDroppedByRelabel:   metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_relabel_metrics_dropped_total{path=%q,url=%q}`, queuePath, sanitizedURL)),
 
-		rowsPushedAfterRelabel: metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_rows_pushed_after_relabel_total{path=%q, url=%q}`, queuePath, sanitizedURL)),
-		rowsDroppedByRelabel:   metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_relabel_metrics_dropped_total{path=%q, url=%q}`, queuePath, sanitizedURL)),
-
-		pushFailures:             metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_push_failures_total{path=%q, url=%q}`, queuePath, sanitizedURL)),
-		rowsDroppedOnPushFailure: metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_samples_dropped_total{path=%q, url=%q}`, queuePath, sanitizedURL)),
-	}
-
-	// Initialize sas
-	sasFile, sasOpts := getStreamAggrOpts(argIdx)
-	if sasFile != "" {
-		sas, err := newStreamAggrConfig(argIdx, rwctx.pushInternalTrackDropped)
-		if err != nil {
-			logger.Fatalf("cannot initialize stream aggregators from -remoteWrite.streamAggr.config=%q: %s", sasFile, err)
-		}
-		rwctx.sas.Store(sas)
-		rwctx.streamAggrKeepInput = streamAggrKeepInput.GetOptionalArg(argIdx)
-		rwctx.streamAggrDropInput = streamAggrDropInput.GetOptionalArg(argIdx)
-		metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, sasFile)).Set(1)
-		metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_success_timestamp_seconds{path=%q}`, sasFile)).Set(fasttime.UnixTimestamp())
-	} else if sasOpts.DedupInterval > 0 {
-		rwctx.deduplicator = streamaggr.NewDeduplicator(rwctx.pushInternalTrackDropped, sasOpts.DedupInterval, sasOpts.DropInputLabels)
+		pushFailures:             metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_push_failures_total{path=%q,url=%q}`, queuePath, sanitizedURL)),
+		rowsDroppedOnPushFailure: metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_samples_dropped_total{path=%q,url=%q}`, queuePath, sanitizedURL)),
 	}
+	rwctx.initStreamAggrConfig()
 
 	return rwctx
 }
 
 func (rwctx *remoteWriteCtx) MustStop() {
 	// sas and deduplicator must be stopped before rwctx is closed
-	// because sas can write pending series to rwctx.pss if there are any
+	// because they can write pending series to rwctx.pss if there are any
 	sas := rwctx.sas.Swap(nil)
 	sas.MustStop()
 
@@ -875,12 +875,20 @@ func (rwctx *remoteWriteCtx) MustStop() {
 
 // TryPush sends tss series to the configured remote write endpoint
 //
-// TryPush can be called concurrently for multiple remoteWriteCtx,
-// so it shouldn't modify tss entries.
+// TryPush doesn't modify tss, so tss can be passed concurrently to TryPush across distinct rwctx instances.
 func (rwctx *remoteWriteCtx) TryPush(tss []prompbmarshal.TimeSeries, forceDropSamplesOnFailure bool) bool {
-	// Apply relabeling
 	var rctx *relabelCtx
 	var v *[]prompbmarshal.TimeSeries
+	defer func() {
+		if rctx == nil {
+			return
+		}
+		*v = prompbmarshal.ResetTimeSeries(tss)
+		tssPool.Put(v)
+		putRelabelCtx(rctx)
+	}()
+
+	// Apply relabeling
 	rcs := allRelabelConfigs.Load()
 	pcs := rcs.perURL[rwctx.idx]
 	if pcs.Len() > 0 {
@@ -916,28 +924,22 @@ func (rwctx *remoteWriteCtx) TryPush(tss []prompbmarshal.TimeSeries, forceDropSa
 		matchIdxsPool.Put(matchIdxs)
 	} else if rwctx.deduplicator != nil {
 		rwctx.deduplicator.Push(tss)
-		tss = tss[:0]
+		return true
 	}
 
-	// Try pushing the data to remote storage
-	ok := rwctx.tryPushInternal(tss)
-
-	// Return back relabeling contexts to the pool
-	if rctx != nil {
-		*v = prompbmarshal.ResetTimeSeries(tss)
-		tssPool.Put(v)
-		putRelabelCtx(rctx)
+	// Try pushing tss to remote storage
+	if rwctx.tryPushInternal(tss) {
+		return true
 	}
 
-	if !ok {
-		rwctx.pushFailures.Inc()
-		if forceDropSamplesOnFailure || rwctx.dropSamplesOnOverload {
-			rwctx.rowsDroppedOnPushFailure.Add(len(tss))
-			return true
-		}
+	// Couldn't push tss to remote storage
+	rwctx.pushFailures.Inc()
+	if forceDropSamplesOnFailure {
+		rowsCount := getRowsCount(tss)
+		rwctx.rowsDroppedOnPushFailure.Add(rowsCount)
+		return true
 	}
-
-	return ok
+	return false
 }
 
 var matchIdxsPool bytesutil.ByteBufferPool
@@ -961,19 +963,26 @@ func (rwctx *remoteWriteCtx) pushInternalTrackDropped(tss []prompbmarshal.TimeSe
 	if rwctx.tryPushInternal(tss) {
 		return
 	}
-	if !rwctx.disableOnDiskQueue {
+	if !rwctx.fq.IsPersistentQueueDisabled() {
 		logger.Panicf("BUG: tryPushInternal must return true if -remoteWrite.disableOnDiskQueue isn't set")
 	}
 	rwctx.pushFailures.Inc()
-	if dropSamplesOnOverload.GetOptionalArg(rwctx.idx) {
-		rowsCount := getRowsCount(tss)
-		rwctx.rowsDroppedOnPushFailure.Add(rowsCount)
-	}
+	rowsCount := getRowsCount(tss)
+	rwctx.rowsDroppedOnPushFailure.Add(rowsCount)
 }
 
 func (rwctx *remoteWriteCtx) tryPushInternal(tss []prompbmarshal.TimeSeries) bool {
 	var rctx *relabelCtx
 	var v *[]prompbmarshal.TimeSeries
+	defer func() {
+		if rctx == nil {
+			return
+		}
+		*v = prompbmarshal.ResetTimeSeries(tss)
+		tssPool.Put(v)
+		putRelabelCtx(rctx)
+	}()
+
 	if len(labelsGlobal) > 0 {
 		// Make a copy of tss before adding extra labels in order to prevent
 		// from affecting time series for other remoteWrite.url configs.
@@ -986,19 +995,11 @@ func (rwctx *remoteWriteCtx) tryPushInternal(tss []prompbmarshal.TimeSeries) boo
 	pss := rwctx.pss
 	idx := rwctx.pssNextIdx.Add(1) % uint64(len(pss))
 
-	ok := pss[idx].TryPush(tss)
-
-	if rctx != nil {
-		*v = prompbmarshal.ResetTimeSeries(tss)
-		tssPool.Put(v)
-		putRelabelCtx(rctx)
-	}
-
-	return ok
+	return pss[idx].TryPush(tss)
 }
 
 var tssPool = &sync.Pool{
-	New: func() interface{} {
+	New: func() any {
 		a := []prompbmarshal.TimeSeries{}
 		return &a
 	},

app/vmagent/remotewrite/remotewrite_test.go

@@ -3,14 +3,12 @@ package remotewrite
 import (
 	"fmt"
 	"math"
-	"os"
 	"reflect"
 	"testing"
 	"time"
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promrelabel"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/prometheus"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/streamaggr"
 	"github.com/VictoriaMetrics/metrics"
 )
@@ -79,24 +77,28 @@ func TestRemoteWriteContext_TryPush_ImmutableTimeseries(t *testing.T) {
 			rowsDroppedByRelabel:   metrics.GetOrCreateCounter(`bar`),
 		}
 		if dedupInterval > 0 {
-			rwctx.deduplicator = streamaggr.NewDeduplicator(nil, dedupInterval, nil)
+			rwctx.deduplicator = streamaggr.NewDeduplicator(nil, dedupInterval, nil, "dedup-global")
 		}
 
-		if len(streamAggrConfig) > 0 {
-			f := createFile(t, []byte(streamAggrConfig))
-			sas, err := streamaggr.LoadFromFile(f.Name(), nil, nil)
+		if streamAggrConfig != "" {
+			pushNoop := func(_ []prompbmarshal.TimeSeries) {}
+			sas, err := streamaggr.LoadFromData([]byte(streamAggrConfig), pushNoop, nil, "global")
 			if err != nil {
 				t.Fatalf("cannot load streamaggr configs: %s", err)
 			}
+			defer sas.MustStop()
 			rwctx.sas.Store(sas)
 		}
 
-		inputTss := mustParsePromMetrics(input)
+		offsetMsecs := time.Now().UnixMilli()
+		inputTss := prompbmarshal.MustParsePromMetrics(input, offsetMsecs)
 		expectedTss := make([]prompbmarshal.TimeSeries, len(inputTss))
 
 		// copy inputTss to make sure it is not mutated during TryPush call
 		copy(expectedTss, inputTss)
-		rwctx.TryPush(inputTss, false)
+		if !rwctx.TryPush(inputTss, false) {
+			t.Fatalf("cannot push samples to rwctx")
+		}
 
 		if !reflect.DeepEqual(expectedTss, inputTss) {
 			t.Fatalf("unexpected samples;\ngot\n%v\nwant\n%v", inputTss, expectedTss)
@@ -165,51 +167,3 @@ metric{env="dev"} 15
 metric{env="bar"} 25
 `)
 }
-
-func mustParsePromMetrics(s string) []prompbmarshal.TimeSeries {
-	var rows prometheus.Rows
-	errLogger := func(s string) {
-		panic(fmt.Errorf("unexpected error when parsing Prometheus metrics: %s", s))
-	}
-	rows.UnmarshalWithErrLogger(s, errLogger)
-	var tss []prompbmarshal.TimeSeries
-	samples := make([]prompbmarshal.Sample, 0, len(rows.Rows))
-	for _, row := range rows.Rows {
-		labels := make([]prompbmarshal.Label, 0, len(row.Tags)+1)
-		labels = append(labels, prompbmarshal.Label{
-			Name:  "__name__",
-			Value: row.Metric,
-		})
-		for _, tag := range row.Tags {
-			labels = append(labels, prompbmarshal.Label{
-				Name:  tag.Key,
-				Value: tag.Value,
-			})
-		}
-		samples = append(samples, prompbmarshal.Sample{
-			Value:     row.Value,
-			Timestamp: row.Timestamp,
-		})
-		ts := prompbmarshal.TimeSeries{
-			Labels:  labels,
-			Samples: samples[len(samples)-1:],
-		}
-		tss = append(tss, ts)
-	}
-	return tss
-}
-
-func createFile(t *testing.T, data []byte) *os.File {
-	t.Helper()
-	f, err := os.CreateTemp("", "")
-	if err != nil {
-		t.Fatal(err)
-	}
-	if err := os.WriteFile(f.Name(), data, 0644); err != nil {
-		t.Fatal(err)
-	}
-	if err := f.Sync(); err != nil {
-		t.Fatal(err)
-	}
-	return f
-}

app/vmagent/remotewrite/streamaggr.go

@@ -36,125 +36,205 @@ var (
 		"before stream de-duplication and aggregation . See https://docs.victoriametrics.com/stream-aggregation/#dropping-unneeded-labels")
 
 	// Per URL config
-	streamAggrConfig = flagutil.NewArrayString("remoteWrite.streamAggr.config", "Optional path to file with stream aggregation config. "+
+	streamAggrConfig = flagutil.NewArrayString("remoteWrite.streamAggr.config", "Optional path to file with stream aggregation config for the corresponding -remoteWrite.url. "+
 		"See https://docs.victoriametrics.com/stream-aggregation/ . "+
 		"See also -remoteWrite.streamAggr.keepInput, -remoteWrite.streamAggr.dropInput and -remoteWrite.streamAggr.dedupInterval")
 	streamAggrDropInput = flagutil.NewArrayBool("remoteWrite.streamAggr.dropInput", "Whether to drop all the input samples after the aggregation "+
-		"with -remoteWrite.streamAggr.config. By default, only aggregates samples are dropped, while the remaining samples "+
+		"with -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. By default, only aggregates samples are dropped, while the remaining samples "+
 		"are written to the corresponding -remoteWrite.url . See also -remoteWrite.streamAggr.keepInput and https://docs.victoriametrics.com/stream-aggregation/")
 	streamAggrKeepInput = flagutil.NewArrayBool("remoteWrite.streamAggr.keepInput", "Whether to keep all the input samples after the aggregation "+
-		"with -remoteWrite.streamAggr.config. By default, only aggregates samples are dropped, while the remaining samples "+
+		"with -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. By default, only aggregates samples are dropped, while the remaining samples "+
 		"are written to the corresponding -remoteWrite.url . See also -remoteWrite.streamAggr.dropInput and https://docs.victoriametrics.com/stream-aggregation/")
 	streamAggrDedupInterval = flagutil.NewArrayDuration("remoteWrite.streamAggr.dedupInterval", 0, "Input samples are de-duplicated with this interval before optional aggregation "+
-		"with -remoteWrite.streamAggr.config . See also -dedup.minScrapeInterval and https://docs.victoriametrics.com/stream-aggregation/#deduplication")
+		"with -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. See also -dedup.minScrapeInterval and https://docs.victoriametrics.com/stream-aggregation/#deduplication")
 	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 . "+
+		"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. Increase this value if "+
-		"you observe incorrect aggregation results after vmagent restarts. It could be caused by receiving unordered delayed data from clients pushing data into the vmagent. "+
+	streamAggrIgnoreFirstIntervals = flag.Int("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")
 	streamAggrDropInputLabels = flagutil.NewArrayString("remoteWrite.streamAggr.dropInputLabels", "An optional list of labels to drop from samples "+
-		"before stream de-duplication and aggregation . See https://docs.victoriametrics.com/stream-aggregation/#dropping-unneeded-labels")
+		"before stream de-duplication and aggregation with -remoteWrite.streamAggr.config and -remoteWrite.streamAggr.dedupInterval at the corresponding -remoteWrite.url. "+
+		"See https://docs.victoriametrics.com/stream-aggregation/#dropping-unneeded-labels")
 )
 
 // CheckStreamAggrConfigs checks -remoteWrite.streamAggr.config and -streamAggr.config.
 func CheckStreamAggrConfigs() error {
-	pushNoop := func(_ []prompbmarshal.TimeSeries) {}
+	// Check global config
+	sas, err := newStreamAggrConfigGlobal()
+	if err != nil {
+		return err
+	}
+	sas.MustStop()
 
-	if _, err := newStreamAggrConfig(-1, pushNoop); err != nil {
-		return fmt.Errorf("could not load -streamAggr.config stream aggregation config: %w", err)
-	}
 	if len(*streamAggrConfig) > len(*remoteWriteURLs) {
-		return fmt.Errorf("too many -remoteWrite.streamAggr.config args: %d; it mustn't exceed the number of -remoteWrite.url args: %d",
-			len(*streamAggrConfig), len(*remoteWriteURLs))
+		return fmt.Errorf("too many -remoteWrite.streamAggr.config args: %d; it mustn't exceed the number of -remoteWrite.url args: %d", len(*streamAggrConfig), len(*remoteWriteURLs))
 	}
+
+	pushNoop := func(_ []prompbmarshal.TimeSeries) {}
 	for idx := range *streamAggrConfig {
-		if _, err := newStreamAggrConfig(idx, pushNoop); err != nil {
+		sas, err := newStreamAggrConfigPerURL(idx, pushNoop)
+		if err != nil {
 			return err
 		}
+		sas.MustStop()
 	}
 	return nil
 }
 
-// HasAnyStreamAggrConfigured checks if any streaming aggregation config provided
-func HasAnyStreamAggrConfigured() bool {
-	return len(*streamAggrConfig) > 0 || *streamAggrGlobalConfig != ""
-}
-
 func reloadStreamAggrConfigs() {
-	reloadStreamAggrConfig(-1, pushToRemoteStoragesDropFailed)
-	for idx, rwctx := range rwctxs {
-		reloadStreamAggrConfig(idx, rwctx.pushInternalTrackDropped)
+	reloadStreamAggrConfigGlobal()
+	for _, rwctx := range rwctxsGlobal {
+		rwctx.reloadStreamAggrConfig()
 	}
 }
 
-func reloadStreamAggrConfig(idx int, pushFunc streamaggr.PushFunc) {
-	path, opts := getStreamAggrOpts(idx)
-	logger.Infof("reloading stream aggregation configs pointed by -remoteWrite.streamAggr.config=%q", path)
-	metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reloads_total{path=%q}`, path)).Inc()
-
-	sasNew, err := newStreamAggrConfigWithOpts(pushFunc, path, opts)
-	if err != nil {
-		metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reloads_errors_total{path=%q}`, path)).Inc()
-		metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, path)).Set(0)
-		logger.Errorf("cannot reload stream aggregation config at %q; continue using the previously loaded config; error: %s", path, err)
+func reloadStreamAggrConfigGlobal() {
+	path := *streamAggrGlobalConfig
+	if path == "" {
 		return
 	}
 
-	var sas *streamaggr.Aggregators
-	if idx < 0 {
-		sas = sasGlobal.Load()
-	} else {
-		sas = rwctxs[idx].sas.Load()
+	logger.Infof("reloading stream aggregation configs pointed by -streamAggr.config=%q", path)
+	metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reloads_total{path=%q}`, path)).Inc()
+
+	sasNew, err := newStreamAggrConfigGlobal()
+	if err != nil {
+		metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reloads_errors_total{path=%q}`, path)).Inc()
+		metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, path)).Set(0)
+		logger.Errorf("cannot reload -streamAggr.config=%q; continue using the previously loaded config; error: %s", path, err)
+		return
 	}
 
+	sas := sasGlobal.Load()
 	if !sasNew.Equal(sas) {
-		var sasOld *streamaggr.Aggregators
-		if idx < 0 {
-			sasOld = sasGlobal.Swap(sasNew)
-		} else {
-			sasOld = rwctxs[idx].sas.Swap(sasNew)
-		}
+		sasOld := sasGlobal.Swap(sasNew)
 		sasOld.MustStop()
-		logger.Infof("successfully reloaded stream aggregation configs at %q", path)
+		logger.Infof("successfully reloaded -streamAggr.config=%q", path)
 	} else {
 		sasNew.MustStop()
-		logger.Infof("successfully reloaded stream aggregation configs at %q", path)
+		logger.Infof("-streamAggr.config=%q wasn't changed since the last reload", path)
 	}
 	metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, path)).Set(1)
 	metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_success_timestamp_seconds{path=%q}`, path)).Set(fasttime.UnixTimestamp())
 }
 
-func getStreamAggrOpts(idx int) (string, *streamaggr.Options) {
-	if idx < 0 {
-		return *streamAggrGlobalConfig, &streamaggr.Options{
-			DedupInterval:        streamAggrGlobalDedupInterval.Duration(),
-			DropInputLabels:      *streamAggrGlobalDropInputLabels,
-			IgnoreOldSamples:     *streamAggrGlobalIgnoreOldSamples,
-			IgnoreFirstIntervals: *streamAggrGlobalIgnoreFirstIntervals,
+func initStreamAggrConfigGlobal() {
+	sas, err := newStreamAggrConfigGlobal()
+	if err != nil {
+		logger.Fatalf("cannot initialize gloabl stream aggregators: %s", err)
+	}
+	if sas != nil {
+		filePath := sas.FilePath()
+		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")
 		}
 	}
-	opts := streamaggr.Options{
+}
+
+func (rwctx *remoteWriteCtx) initStreamAggrConfig() {
+	idx := rwctx.idx
+
+	sas, err := rwctx.newStreamAggrConfig()
+	if err != nil {
+		logger.Fatalf("cannot initialize stream aggregators: %s", err)
+	}
+	if sas != nil {
+		filePath := sas.FilePath()
+		rwctx.sas.Store(sas)
+		rwctx.streamAggrKeepInput = streamAggrKeepInput.GetOptionalArg(idx)
+		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)
+		}
+	}
+}
+
+func (rwctx *remoteWriteCtx) reloadStreamAggrConfig() {
+	path := streamAggrConfig.GetOptionalArg(rwctx.idx)
+	if path == "" {
+		return
+	}
+
+	logger.Infof("reloading stream aggregation configs pointed by -remoteWrite.streamAggr.config=%q", path)
+	metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reloads_total{path=%q}`, path)).Inc()
+
+	sasNew, err := rwctx.newStreamAggrConfig()
+	if err != nil {
+		metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reloads_errors_total{path=%q}`, path)).Inc()
+		metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, path)).Set(0)
+		logger.Errorf("cannot reload -remoteWrite.streamAggr.config=%q; continue using the previously loaded config; error: %s", path, err)
+		return
+	}
+
+	sas := rwctx.sas.Load()
+	if !sasNew.Equal(sas) {
+		sasOld := rwctx.sas.Swap(sasNew)
+		sasOld.MustStop()
+		logger.Infof("successfully reloaded -remoteWrite.streamAggr.config=%q", path)
+	} else {
+		sasNew.MustStop()
+		logger.Infof("-remoteWrite.streamAggr.config=%q wasn't changed since the last reload", path)
+	}
+	metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, path)).Set(1)
+	metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_success_timestamp_seconds{path=%q}`, path)).Set(fasttime.UnixTimestamp())
+}
+
+func newStreamAggrConfigGlobal() (*streamaggr.Aggregators, error) {
+	path := *streamAggrGlobalConfig
+	if path == "" {
+		return nil, nil
+	}
+
+	opts := &streamaggr.Options{
+		DedupInterval:        streamAggrGlobalDedupInterval.Duration(),
+		DropInputLabels:      *streamAggrGlobalDropInputLabels,
+		IgnoreOldSamples:     *streamAggrGlobalIgnoreOldSamples,
+		IgnoreFirstIntervals: *streamAggrGlobalIgnoreFirstIntervals,
+	}
+
+	sas, err := streamaggr.LoadFromFile(path, pushToRemoteStoragesTrackDropped, opts, "global")
+	if err != nil {
+		return nil, fmt.Errorf("cannot load -streamAggr.config=%q: %w", *streamAggrGlobalConfig, err)
+	}
+	return sas, nil
+}
+
+func (rwctx *remoteWriteCtx) newStreamAggrConfig() (*streamaggr.Aggregators, error) {
+	return newStreamAggrConfigPerURL(rwctx.idx, rwctx.pushInternalTrackDropped)
+}
+
+func newStreamAggrConfigPerURL(idx int, pushFunc streamaggr.PushFunc) (*streamaggr.Aggregators, error) {
+	path := streamAggrConfig.GetOptionalArg(idx)
+	if path == "" {
+		return nil, nil
+	}
+
+	alias := fmt.Sprintf("%d:secret-url", idx+1)
+	if *showRemoteWriteURL {
+		alias = fmt.Sprintf("%d:%s", idx+1, remoteWriteURLs.GetOptionalArg(idx))
+	}
+	opts := &streamaggr.Options{
 		DedupInterval:        streamAggrDedupInterval.GetOptionalArg(idx),
 		DropInputLabels:      *streamAggrDropInputLabels,
 		IgnoreOldSamples:     streamAggrIgnoreOldSamples.GetOptionalArg(idx),
 		IgnoreFirstIntervals: *streamAggrIgnoreFirstIntervals,
 	}
-	if len(*streamAggrConfig) == 0 {
-		return "", &opts
-	}
-	return streamAggrConfig.GetOptionalArg(idx), &opts
-}
 
-func newStreamAggrConfigWithOpts(pushFunc streamaggr.PushFunc, path string, opts *streamaggr.Options) (*streamaggr.Aggregators, error) {
-	if len(path) == 0 {
-		// Skip empty stream aggregation config.
-		return nil, nil
+	sas, err := streamaggr.LoadFromFile(path, pushFunc, opts, alias)
+	if err != nil {
+		return nil, fmt.Errorf("cannot load -remoteWrite.streamAggr.config=%q: %w", path, err)
 	}
-	return streamaggr.LoadFromFile(path, pushFunc, opts)
-}
-
-func newStreamAggrConfig(idx int, pushFunc streamaggr.PushFunc) (*streamaggr.Aggregators, error) {
-	path, opts := getStreamAggrOpts(idx)
-	return newStreamAggrConfigWithOpts(pushFunc, path, opts)
+	return sas, nil
 }

app/vmagent/statsd/request_handler.go

@@ -1,72 +0,0 @@
-package statsd
-
-import (
-	"io"
-
-	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/common"
-	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/remotewrite"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/auth"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
-	parser "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/statsd"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/statsd/stream"
-	"github.com/VictoriaMetrics/metrics"
-)
-
-var (
-	rowsInserted  = metrics.NewCounter(`vmagent_rows_inserted_total{type="statsd"}`)
-	rowsPerInsert = metrics.NewHistogram(`vmagent_rows_per_insert{type="statsd"}`)
-)
-
-// InsertHandler processes remote write for statsd plaintext protocol.
-//
-// See https://github.com/statsd/statsd/blob/master/docs/metric_types.md
-func InsertHandler(r io.Reader) error {
-	return stream.Parse(r, false, func(rows []parser.Row) error {
-		return insertRows(nil, rows)
-	})
-}
-
-func insertRows(at *auth.Token, rows []parser.Row) error {
-	ctx := common.GetPushCtx()
-	defer common.PutPushCtx(ctx)
-
-	tssDst := ctx.WriteRequest.Timeseries[:0]
-	labels := ctx.Labels[:0]
-	samples := ctx.Samples[:0]
-	for i := range rows {
-		r := &rows[i]
-		labelsLen := len(labels)
-		labels = append(labels, prompbmarshal.Label{
-			Name:  "__name__",
-			Value: r.Metric,
-		})
-		for j := range r.Tags {
-			tag := &r.Tags[j]
-			labels = append(labels, prompbmarshal.Label{
-				Name:  tag.Key,
-				Value: tag.Value,
-			})
-		}
-		samplesLen := len(samples)
-		for _, v := range r.Values {
-			samples = append(samples, prompbmarshal.Sample{
-				Value:     v,
-				Timestamp: r.Timestamp,
-			})
-		}
-
-		tssDst = append(tssDst, prompbmarshal.TimeSeries{
-			Labels:  labels[labelsLen:],
-			Samples: samples[samplesLen:],
-		})
-	}
-	ctx.WriteRequest.Timeseries = tssDst
-	ctx.Labels = labels
-	ctx.Samples = samples
-	if !remotewrite.TryPush(at, &ctx.WriteRequest) {
-		return remotewrite.ErrQueueFullHTTPRetry
-	}
-	rowsInserted.Add(len(rows))
-	rowsPerInsert.Update(float64(len(rows)))
-	return nil
-}

app/vmalert-tool/Makefile

@@ -81,6 +81,9 @@ vmalert-tool-linux-ppc64le:
 vmalert-tool-linux-s390x:
 	APP_NAME=vmalert-tool CGO_ENABLED=0 GOOS=linux GOARCH=s390x $(MAKE) app-local-goos-goarch
 
+vmalert-tool-linux-loong64:
+	APP_NAME=vmalert-tool CGO_ENABLED=0 GOOS=linux GOARCH=loong64 $(MAKE) app-local-goos-goarch
+
 vmalert-tool-linux-386:
 	APP_NAME=vmalert-tool CGO_ENABLED=0 GOOS=linux GOARCH=386 $(MAKE) app-local-goos-goarch
 

app/vmalert-tool/main.go

@@ -26,8 +26,14 @@ func main() {
 				UsageText: "More info in https://docs.victoriametrics.com/vmalert-tool.html#Unit-testing-for-rules",
 				Flags: []cli.Flag{
 					&cli.StringSliceFlag{
-						Name:     "files",
-						Usage:    "files to run unittest with. Supports an array of values separated by comma or specified via multiple flags.",
+						Name: "files",
+						Usage: `File path or http url with test files. Supports an array of values separated by comma or specified via multiple flags.
+						Supports hierarchical patterns and regexpes.
+Examples:
+ -files="/path/to/file". Path to a single test file.
+ -files="http://<some-server-addr>/path/to/test.yaml". HTTP URL to a test file.
+ -files="dir/**/*.yaml". Includes all the .yaml files in "dir" subfolders recursively.
+ `,
 						Required: true,
 					},
 					&cli.BoolFlag{

app/vmalert-tool/multiarch/Dockerfile

@@ -1,7 +1,7 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
 ARG certs_image
 ARG root_image
-FROM $certs_image as certs
+FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 
 FROM $root_image

app/vmalert-tool/unittest/input.go

@@ -74,10 +74,7 @@ func writeInputSeries(input []series, interval *promutils.Duration, startStamp t
 		r.Timeseries = append(r.Timeseries, testutil.TimeSeries{Labels: ls, Samples: samples})
 	}
 
-	data, err := testutil.Compress(r)
-	if err != nil {
-		return fmt.Errorf("failed to compress data: %v", err)
-	}
+	data := testutil.Compress(r)
 	// write input series to vm
 	httpWrite(dst, bytes.NewBuffer(data))
 	vmstorage.Storage.DebugFlush()

app/vmalert-tool/unittest/input_test.go

@@ -6,88 +6,61 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/decimal"
 )
 
-func TestParseInputValue(t *testing.T) {
-	testCases := []struct {
-		input  string
-		exp    []sequenceValue
-		failed bool
-	}{
-		{
-			"",
-			nil,
-			true,
-		},
-		{
-			"testfailed",
-			nil,
-			true,
-		},
-		// stale doesn't support operations
-		{
-			"stalex3",
-			nil,
-			true,
-		},
-		{
-			"-4",
-			[]sequenceValue{{Value: -4}},
-			false,
-		},
-		{
-			"_",
-			[]sequenceValue{{Omitted: true}},
-			false,
-		},
-		{
-			"stale",
-			[]sequenceValue{{Value: decimal.StaleNaN}},
-			false,
-		},
-		{
-			"-4x1",
-			[]sequenceValue{{Value: -4}, {Value: -4}},
-			false,
-		},
-		{
-			"_x1",
-			[]sequenceValue{{Omitted: true}},
-			false,
-		},
-		{
-			"1+1x2 0.1 0.1+0.3x2 3.14",
-			[]sequenceValue{{Value: 1}, {Value: 2}, {Value: 3}, {Value: 0.1}, {Value: 0.1}, {Value: 0.4}, {Value: 0.7}, {Value: 3.14}},
-			false,
-		},
-		{
-			"2-1x4",
-			[]sequenceValue{{Value: 2}, {Value: 1}, {Value: 0}, {Value: -1}, {Value: -2}},
-			false,
-		},
-		{
-			"1+1x1 _ -4 stale 3+20x1",
-			[]sequenceValue{{Value: 1}, {Value: 2}, {Omitted: true}, {Value: -4}, {Value: decimal.StaleNaN}, {Value: 3}, {Value: 23}},
-			false,
-		},
+func TestParseInputValue_Failure(t *testing.T) {
+	f := func(input string) {
+		t.Helper()
+
+		_, err := parseInputValue(input, true)
+		if err == nil {
+			t.Fatalf("expecting non-nil error")
+		}
 	}
 
-	for _, tc := range testCases {
-		output, err := parseInputValue(tc.input, true)
-		if err != nil != tc.failed {
-			t.Fatalf("failed to parse %s, expect %t, got %t", tc.input, tc.failed, err != nil)
+	f("")
+	f("testfailed")
+
+	// stale doesn't support operations
+	f("stalex3")
+}
+
+func TestParseInputValue_Success(t *testing.T) {
+	f := func(input string, outputExpected []sequenceValue) {
+		t.Helper()
+
+		output, err := parseInputValue(input, true)
+		if err != nil {
+			t.Fatalf("unexpected error in parseInputValue: %s", err)
 		}
-		if len(tc.exp) != len(output) {
-			t.Fatalf("expect %v, got %v", tc.exp, output)
+
+		if len(outputExpected) != len(output) {
+			t.Fatalf("unexpected output length; got %d; want %d", len(outputExpected), len(output))
 		}
-		for i := 0; i < len(tc.exp); i++ {
-			if tc.exp[i].Omitted != output[i].Omitted {
-				t.Fatalf("expect %v, got %v", tc.exp, output)
+		for i := 0; i < len(outputExpected); i++ {
+			if outputExpected[i].Omitted != output[i].Omitted {
+				t.Fatalf("unexpected Omitted field in the output\ngot\n%v\nwant\n%v", output, outputExpected)
 			}
-			if tc.exp[i].Value != output[i].Value {
-				if decimal.IsStaleNaN(tc.exp[i].Value) && decimal.IsStaleNaN(output[i].Value) {
+			if outputExpected[i].Value != output[i].Value {
+				if decimal.IsStaleNaN(outputExpected[i].Value) && decimal.IsStaleNaN(output[i].Value) {
 					continue
 				}
-				t.Fatalf("expect %v, got %v", tc.exp, output)
+				t.Fatalf("unexpeccted Value field in the output\ngot\n%v\nwant\n%v", output, outputExpected)
 			}
 		}
 	}
+
+	f("-4", []sequenceValue{{Value: -4}})
+
+	f("_", []sequenceValue{{Omitted: true}})
+
+	f("stale", []sequenceValue{{Value: decimal.StaleNaN}})
+
+	f("-4x1", []sequenceValue{{Value: -4}, {Value: -4}})
+
+	f("_x1", []sequenceValue{{Omitted: true}})
+
+	f("1+1x2 0.1 0.1+0.3x2 3.14", []sequenceValue{{Value: 1}, {Value: 2}, {Value: 3}, {Value: 0.1}, {Value: 0.1}, {Value: 0.4}, {Value: 0.7}, {Value: 3.14}})
+
+	f("2-1x4", []sequenceValue{{Value: 2}, {Value: 1}, {Value: 0}, {Value: -1}, {Value: -2}})
+
+	f("1+1x1 _ -4 stale 3+20x1", []sequenceValue{{Value: 1}, {Value: 2}, {Omitted: true}, {Value: -4}, {Value: decimal.StaleNaN}, {Value: 3}, {Value: 23}})
 }

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

@@ -37,3 +37,6 @@ groups:
     rules:
       - record: t3
         expr: t1
+
+  - name: emptyGroup
+    rules: []

app/vmalert-tool/unittest/unittest.go

@@ -9,10 +9,12 @@ import (
 	"path/filepath"
 	"reflect"
 	"sort"
+	"strings"
 	"time"
 
 	"gopkg.in/yaml.v2"
 
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/config"
 	vmalertconfig "github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/config"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/datasource"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/notifier"
@@ -66,36 +68,38 @@ func UnitTest(files []string, disableGroupLabel bool) bool {
 	defer vminsert.Stop()
 	defer vmselect.Stop()
 	disableAlertgroupLabel = disableGroupLabel
-	return rulesUnitTest(files)
-}
 
-func rulesUnitTest(files []string) bool {
+	testfiles, err := config.ReadFromFS(files)
+	if err != nil {
+		fmt.Println("  FAILED")
+		fmt.Printf("\nfailed to read test files: \n%v", err)
+	}
 	var failed bool
-	for _, f := range files {
-		if err := ruleUnitTest(f); err != nil {
+	for fileName, file := range testfiles {
+		if err := ruleUnitTest(fileName, file); err != nil {
 			fmt.Println("  FAILED")
-			fmt.Printf("\nfailed to run unit test for file %q: \n%v", f, err)
+			fmt.Printf("\nfailed to run unit test for file %q: \n%v", file, err)
 			failed = true
 		} else {
 			fmt.Println("  SUCCESS")
 		}
 	}
+
 	return failed
 }
 
-func ruleUnitTest(filename string) []error {
+func ruleUnitTest(filename string, content []byte) []error {
 	fmt.Println("\nUnit Testing: ", filename)
-	b, err := os.ReadFile(filename)
-	if err != nil {
-		return []error{fmt.Errorf("failed to read file: %w", err)}
-	}
-
 	var unitTestInp unitTestFile
-	if err := yaml.UnmarshalStrict(b, &unitTestInp); err != nil {
+	if err := yaml.UnmarshalStrict(content, &unitTestInp); err != nil {
 		return []error{fmt.Errorf("failed to unmarshal file: %w", err)}
 	}
-	if err := resolveAndGlobFilepaths(filepath.Dir(filename), &unitTestInp); err != nil {
-		return []error{fmt.Errorf("failed to resolve path for `rule_files`: %w", err)}
+
+	// add file directory for rule files if needed
+	for i, rf := range unitTestInp.RuleFiles {
+		if rf != "" && !filepath.IsAbs(rf) && !strings.HasPrefix(rf, "http") {
+			unitTestInp.RuleFiles[i] = filepath.Join(filepath.Dir(filename), rf)
+		}
 	}
 
 	if unitTestInp.EvaluationInterval.Duration() == 0 {
@@ -139,24 +143,24 @@ func verifyTestGroup(group testGroup) error {
 	}
 	for _, at := range group.AlertRuleTests {
 		if at.Alertname == "" {
-			return fmt.Errorf("\n%s    missing required filed \"alertname\"", testGroupName)
+			return fmt.Errorf("\n%s    missing required field \"alertname\"", testGroupName)
 		}
 		if !disableAlertgroupLabel && at.GroupName == "" {
-			return fmt.Errorf("\n%s    missing required filed \"groupname\" when flag \"disableAlertGroupLabel\" is false", testGroupName)
+			return fmt.Errorf("\n%s    missing required field \"groupname\" when flag \"disableAlertgroupLabel\" is false", testGroupName)
 		}
 		if disableAlertgroupLabel && at.GroupName != "" {
-			return fmt.Errorf("\n%s    shouldn't set filed \"groupname\" when flag \"disableAlertGroupLabel\" is true", testGroupName)
+			return fmt.Errorf("\n%s    shouldn't set field \"groupname\" when flag \"disableAlertgroupLabel\" is true", testGroupName)
 		}
 		if at.EvalTime == nil {
-			return fmt.Errorf("\n%s    missing required filed \"eval_time\"", testGroupName)
+			return fmt.Errorf("\n%s    missing required field \"eval_time\"", testGroupName)
 		}
 	}
 	for _, et := range group.MetricsqlExprTests {
 		if et.Expr == "" {
-			return fmt.Errorf("\n%s    missing required filed \"expr\"", testGroupName)
+			return fmt.Errorf("\n%s    missing required field \"expr\"", testGroupName)
 		}
 		if et.EvalTime == nil {
-			return fmt.Errorf("\n%s    missing required filed \"eval_time\"", testGroupName)
+			return fmt.Errorf("\n%s    missing required field \"eval_time\"", testGroupName)
 		}
 	}
 	return nil
@@ -235,30 +239,6 @@ func tearDown() {
 	fs.MustRemoveAll(storagePath)
 }
 
-// resolveAndGlobFilepaths joins all relative paths in a configuration
-// with a given base directory and replaces all globs with matching files.
-func resolveAndGlobFilepaths(baseDir string, utf *unitTestFile) error {
-	for i, rf := range utf.RuleFiles {
-		if rf != "" && !filepath.IsAbs(rf) {
-			utf.RuleFiles[i] = filepath.Join(baseDir, rf)
-		}
-	}
-
-	var globbedFiles []string
-	for _, rf := range utf.RuleFiles {
-		m, err := filepath.Glob(rf)
-		if err != nil {
-			return err
-		}
-		if len(m) == 0 {
-			fmt.Fprintln(os.Stderr, "  WARNING: no file match pattern", rf)
-		}
-		globbedFiles = append(globbedFiles, m...)
-	}
-	utf.RuleFiles = globbedFiles
-	return nil
-}
-
 func (tg *testGroup) test(evalInterval time.Duration, groupOrderMap map[string]int, testGroups []vmalertconfig.Group) (checkErrs []error) {
 	// set up vmstorage and http server for ingest and read queries
 	setUp()
@@ -316,11 +296,15 @@ func (tg *testGroup) test(evalInterval time.Duration, groupOrderMap map[string]i
 	maxEvalTime := testStartTime.Add(tg.maxEvalTime())
 	for ts := testStartTime; ts.Before(maxEvalTime) || ts.Equal(maxEvalTime); ts = ts.Add(evalInterval) {
 		for _, g := range groups {
+			if len(g.Rules) == 0 {
+				continue
+			}
 			errs := g.ExecOnce(context.Background(), func() []notifier.Notifier { return nil }, rw, ts)
 			for err := range errs {
 				if err != nil {
 					checkErrs = append(checkErrs, fmt.Errorf("\nfailed to exec group: %q, time: %s, err: %w", g.Name,
 						ts, err))
+					return
 				}
 			}
 			// flush series after each group evaluation
@@ -374,7 +358,7 @@ func (tg *testGroup) test(evalInterval time.Duration, groupOrderMap map[string]i
 						if expAlert.ExpLabels == nil {
 							expAlert.ExpLabels = make(map[string]string)
 						}
-						// alertGroupNameLabel is added as additional labels when `disableAlertGroupLabel` is false
+						// alertGroupNameLabel is added as additional labels when `disableAlertgroupLabel` is false
 						if !disableAlertgroupLabel {
 							expAlert.ExpLabels["alertgroup"] = groupname
 						}

app/vmalert-tool/unittest/unittest_test.go

@@ -14,34 +14,33 @@ func TestMain(m *testing.M) {
 	os.Exit(m.Run())
 }
 
-func TestUnitRule(t *testing.T) {
-	testCases := []struct {
-		name              string
-		disableGroupLabel bool
-		files             []string
-		failed            bool
-	}{
-		{
-			name:   "run multi files",
-			files:  []string{"./testdata/test1.yaml", "./testdata/test2.yaml"},
-			failed: false,
-		},
-		{
-			name:              "disable group label",
-			disableGroupLabel: true,
-			files:             []string{"./testdata/disable-group-label.yaml"},
-			failed:            false,
-		},
-		{
-			name:   "failing test",
-			files:  []string{"./testdata/failed-test.yaml"},
-			failed: true,
-		},
-	}
-	for _, tc := range testCases {
-		fail := UnitTest(tc.files, tc.disableGroupLabel)
-		if fail != tc.failed {
-			t.Fatalf("failed to test %s, expect %t, got %t", tc.name, tc.failed, fail)
+func TestUnitTest_Failure(t *testing.T) {
+	f := func(files []string) {
+		t.Helper()
+
+		failed := UnitTest(files, false)
+		if !failed {
+			t.Fatalf("expecting failed test")
 		}
 	}
+
+	// failing test
+	f([]string{"./testdata/failed-test.yaml"})
+}
+
+func TestUnitTest_Success(t *testing.T) {
+	f := func(disableGroupLabel bool, files []string) {
+		t.Helper()
+
+		failed := UnitTest(files, disableGroupLabel)
+		if failed {
+			t.Fatalf("unexpected failed test")
+		}
+	}
+
+	// run multi files
+	f(false, []string{"./testdata/test1.yaml", "./testdata/test2.yaml"})
+
+	// disable group label
+	f(true, []string{"./testdata/disable-group-label.yaml"})
 }

app/vmalert/Makefile

@@ -101,8 +101,7 @@ replay-vmalert: vmalert
 		-remoteWrite.url=http://localhost:8428 \
 		-external.label=cluster=east-1 \
 		-external.label=replica=a \
-		-replay.timeFrom=2021-05-11T07:21:43Z \
-		-replay.timeTo=2021-05-29T18:40:43Z
+		-replay.timeFrom=2024-06-01T00:00:00Z
 
 vmalert-linux-amd64:
 	APP_NAME=vmalert CGO_ENABLED=1 GOOS=linux GOARCH=amd64 $(MAKE) app-local-goos-goarch

app/vmalert/config/config.go

@@ -45,11 +45,11 @@ type Group struct {
 	// EvalAlignment will make the timestamp of group query requests be aligned with interval
 	EvalAlignment *bool `yaml:"eval_alignment,omitempty"`
 	// Catches all undefined fields and must be empty after parsing.
-	XXX map[string]interface{} `yaml:",inline"`
+	XXX map[string]any `yaml:",inline"`
 }
 
 // UnmarshalYAML implements the yaml.Unmarshaler interface.
-func (g *Group) UnmarshalYAML(unmarshal func(interface{}) error) error {
+func (g *Group) UnmarshalYAML(unmarshal func(any) error) error {
 	type group Group
 	if err := unmarshal((*group)(g)); err != nil {
 		return err
@@ -142,11 +142,11 @@ type Rule struct {
 	UpdateEntriesLimit *int `yaml:"update_entries_limit,omitempty"`
 
 	// Catches all undefined fields and must be empty after parsing.
-	XXX map[string]interface{} `yaml:",inline"`
+	XXX map[string]any `yaml:",inline"`
 }
 
 // UnmarshalYAML implements the yaml.Unmarshaler interface.
-func (r *Rule) UnmarshalYAML(unmarshal func(interface{}) error) error {
+func (r *Rule) UnmarshalYAML(unmarshal func(any) error) error {
 	type rule Rule
 	if err := unmarshal((*rule)(r)); err != nil {
 		return err
@@ -234,7 +234,7 @@ func ParseSilent(pathPatterns []string, validateTplFn ValidateTplFn, validateExp
 	cLogger.Suppress(true)
 	defer cLogger.Suppress(false)
 
-	files, err := readFromFS(pathPatterns)
+	files, err := ReadFromFS(pathPatterns)
 	if err != nil {
 		return nil, fmt.Errorf("failed to read from the config: %w", err)
 	}
@@ -243,7 +243,7 @@ func ParseSilent(pathPatterns []string, validateTplFn ValidateTplFn, validateExp
 
 // Parse parses rule configs from given file patterns
 func Parse(pathPatterns []string, validateTplFn ValidateTplFn, validateExpressions bool) ([]Group, error) {
-	files, err := readFromFS(pathPatterns)
+	files, err := ReadFromFS(pathPatterns)
 	if err != nil {
 		return nil, fmt.Errorf("failed to read from the config: %w", err)
 	}
@@ -301,7 +301,7 @@ func parseConfig(data []byte) ([]Group, error) {
 	g := struct {
 		Groups []Group `yaml:"groups"`
 		// Catches all undefined fields and must be empty after parsing.
-		XXX map[string]interface{} `yaml:",inline"`
+		XXX map[string]any `yaml:",inline"`
 	}{}
 	err = yaml.Unmarshal(data, &g)
 	if err != nil {
@@ -310,7 +310,7 @@ func parseConfig(data []byte) ([]Group, error) {
 	return g.Groups, checkOverflow(g.XXX, "config")
 }
 
-func checkOverflow(m map[string]interface{}, ctx string) error {
+func checkOverflow(m map[string]any, ctx string) error {
 	if len(m) > 0 {
 		var keys []string
 		for k := range m {

app/vmalert/config/config_test.go

@@ -23,12 +23,6 @@ func TestMain(m *testing.M) {
 	os.Exit(m.Run())
 }
 
-func TestParseGood(t *testing.T) {
-	if _, err := Parse([]string{"testdata/rules/*good.rules", "testdata/dir/*good.*"}, notifier.ValidateTemplates, true); err != nil {
-		t.Errorf("error parsing files %s", err)
-	}
-}
-
 func TestParseFromURL(t *testing.T) {
 	mux := http.NewServeMux()
 	mux.HandleFunc("/bad", func(w http.ResponseWriter, _ *http.Request) {
@@ -55,438 +49,353 @@ groups:
 	defer srv.Close()
 
 	if _, err := Parse([]string{srv.URL + "/good-alert", srv.URL + "/good-rr"}, notifier.ValidateTemplates, true); err != nil {
-		t.Errorf("error parsing URLs %s", err)
+		t.Fatalf("error parsing URLs %s", err)
 	}
 
 	if _, err := Parse([]string{srv.URL + "/bad"}, notifier.ValidateTemplates, true); err == nil {
-		t.Errorf("expected parsing error: %s", err)
+		t.Fatalf("expected parsing error: %s", err)
 	}
 }
 
-func TestParseBad(t *testing.T) {
-	testCases := []struct {
-		path   []string
-		expErr string
-	}{
-		{
-			[]string{"testdata/rules/rules_interval_bad.rules"},
-			"eval_offset should be smaller than interval",
-		},
-		{
-			[]string{"testdata/rules/rules0-bad.rules"},
-			"unexpected token",
-		},
-		{
-			[]string{"testdata/dir/rules0-bad.rules"},
-			"error parsing annotation",
-		},
-		{
-			[]string{"testdata/dir/rules1-bad.rules"},
-			"duplicate in file",
-		},
-		{
-			[]string{"testdata/dir/rules2-bad.rules"},
-			"function \"unknown\" not defined",
-		},
-		{
-			[]string{"testdata/dir/rules3-bad.rules"},
-			"either `record` or `alert` must be set",
-		},
-		{
-			[]string{"testdata/dir/rules4-bad.rules"},
-			"either `record` or `alert` must be set",
-		},
-		{
-			[]string{"testdata/rules/rules1-bad.rules"},
-			"bad graphite expr",
-		},
-		{
-			[]string{"testdata/dir/rules6-bad.rules"},
-			"missing ':' in header",
-		},
-		{
-			[]string{"http://unreachable-url"},
-			"failed to",
-		},
+func TestParse_Success(t *testing.T) {
+	_, err := Parse([]string{"testdata/rules/*good.rules", "testdata/dir/*good.*"}, notifier.ValidateTemplates, true)
+	if err != nil {
+		t.Fatalf("error parsing files %s", err)
 	}
-	for _, tc := range testCases {
-		_, err := Parse(tc.path, notifier.ValidateTemplates, true)
+}
+
+func TestParse_Failure(t *testing.T) {
+	f := func(paths []string, errStrExpected string) {
+		t.Helper()
+
+		_, err := Parse(paths, notifier.ValidateTemplates, true)
 		if err == nil {
-			t.Errorf("expected to get error")
-			return
+			t.Fatalf("expected to get error")
 		}
-		if !strings.Contains(err.Error(), tc.expErr) {
-			t.Errorf("expected err to contain %q; got %q instead", tc.expErr, err)
+		if !strings.Contains(err.Error(), errStrExpected) {
+			t.Fatalf("expected err to contain %q; got %q instead", errStrExpected, err)
 		}
 	}
+
+	f([]string{"testdata/rules/rules_interval_bad.rules"}, "eval_offset should be smaller than interval")
+	f([]string{"testdata/rules/rules0-bad.rules"}, "unexpected token")
+	f([]string{"testdata/dir/rules0-bad.rules"}, "error parsing annotation")
+	f([]string{"testdata/dir/rules1-bad.rules"}, "duplicate in file")
+	f([]string{"testdata/dir/rules2-bad.rules"}, "function \"unknown\" not defined")
+	f([]string{"testdata/dir/rules3-bad.rules"}, "either `record` or `alert` must be set")
+	f([]string{"testdata/dir/rules4-bad.rules"}, "either `record` or `alert` must be set")
+	f([]string{"testdata/rules/rules1-bad.rules"}, "bad graphite expr")
+	f([]string{"testdata/dir/rules6-bad.rules"}, "missing ':' in header")
+	f([]string{"http://unreachable-url"}, "failed to")
 }
 
-func TestRule_Validate(t *testing.T) {
+func TestRuleValidate(t *testing.T) {
 	if err := (&Rule{}).Validate(); err == nil {
-		t.Errorf("expected empty name error")
+		t.Fatalf("expected empty name error")
 	}
 	if err := (&Rule{Alert: "alert"}).Validate(); err == nil {
-		t.Errorf("expected empty expr error")
+		t.Fatalf("expected empty expr error")
 	}
 	if err := (&Rule{Alert: "alert", Expr: "test>0"}).Validate(); err != nil {
-		t.Errorf("expected valid rule; got %s", err)
+		t.Fatalf("expected valid rule; got %s", err)
 	}
 }
 
-func TestGroup_Validate(t *testing.T) {
-	testCases := []struct {
-		group               *Group
-		rules               []Rule
-		validateAnnotations bool
-		validateExpressions bool
-		expErr              string
-	}{
-		{
-			group:  &Group{},
-			expErr: "group name must be set",
-		},
-		{
-			group: &Group{
-				Name:     "negative interval",
-				Interval: promutils.NewDuration(-1),
+func TestGroupValidate_Failure(t *testing.T) {
+	f := func(group *Group, validateExpressions bool, errStrExpected string) {
+		t.Helper()
+
+		err := group.Validate(nil, validateExpressions)
+		if err == nil {
+			t.Fatalf("expecting non-nil error")
+		}
+		errStr := err.Error()
+		if !strings.Contains(errStr, errStrExpected) {
+			t.Fatalf("missing %q in the returned error %q", errStrExpected, errStr)
+		}
+	}
+
+	f(&Group{}, false, "group name must be set")
+
+	f(&Group{
+		Name:     "negative interval",
+		Interval: promutils.NewDuration(-1),
+	}, false, "interval shouldn't be lower than 0")
+
+	f(&Group{
+		Name:       "wrong eval_offset",
+		Interval:   promutils.NewDuration(time.Minute),
+		EvalOffset: promutils.NewDuration(2 * time.Minute),
+	}, false, "eval_offset should be smaller than interval")
+
+	f(&Group{
+		Name:  "wrong limit",
+		Limit: -1,
+	}, false, "invalid limit")
+
+	f(&Group{
+		Name:        "wrong concurrency",
+		Concurrency: -1,
+	}, false, "invalid concurrency")
+
+	f(&Group{
+		Name: "test",
+		Rules: []Rule{
+			{
+				Alert: "alert",
+				Expr:  "up == 1",
 			},
-			expErr: "interval shouldn't be lower than 0",
-		},
-		{
-			group: &Group{
-				Name:       "wrong eval_offset",
-				Interval:   promutils.NewDuration(time.Minute),
-				EvalOffset: promutils.NewDuration(2 * time.Minute),
+			{
+				Alert: "alert",
+				Expr:  "up == 1",
 			},
-			expErr: "eval_offset should be smaller than interval",
 		},
-		{
-			group: &Group{
-				Name:  "wrong limit",
-				Limit: -1,
+	}, false, "duplicate")
+
+	f(&Group{
+		Name: "test",
+		Rules: []Rule{
+			{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+				"summary": "{{ value|query }}",
+			}},
+			{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+				"summary": "{{ value|query }}",
+			}},
+		},
+	}, false, "duplicate")
+
+	f(&Group{
+		Name: "test",
+		Rules: []Rule{
+			{Record: "record", Expr: "up == 1", Labels: map[string]string{
+				"summary": "{{ value|query }}",
+			}},
+			{Record: "record", Expr: "up == 1", Labels: map[string]string{
+				"summary": "{{ value|query }}",
+			}},
+		},
+	}, false, "duplicate")
+
+	f(&Group{
+		Name: "test",
+		Rules: []Rule{
+			{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+				"summary": "{{ value|query }}",
+			}},
+			{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+				"description": "{{ value|query }}",
+			}},
+		},
+	}, false, "duplicate")
+
+	f(&Group{
+		Name: "test",
+		Rules: []Rule{
+			{Record: "alert", Expr: "up == 1", Labels: map[string]string{
+				"summary": "{{ value|query }}",
+			}},
+			{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+				"summary": "{{ value|query }}",
+			}},
+		},
+	}, false, "duplicate")
+
+	f(&Group{
+		Name: "test graphite prometheus bad expr",
+		Type: NewGraphiteType(),
+		Rules: []Rule{
+			{
+				Expr: "sum(up == 0 ) by (host)",
+				For:  promutils.NewDuration(10 * time.Millisecond),
 			},
-			expErr: "invalid limit",
-		},
-		{
-			group: &Group{
-				Name:        "wrong concurrency",
-				Concurrency: -1,
+			{
+				Expr: "sumSeries(time('foo.bar',10))",
 			},
-			expErr: "invalid concurrency",
 		},
-		{
-			group: &Group{
-				Name: "test",
-				Rules: []Rule{
-					{
-						Record: "record",
-						Expr:   "up | 0",
-					},
+	}, false, "invalid rule")
+
+	f(&Group{
+		Name: "test graphite inherit",
+		Type: NewGraphiteType(),
+		Rules: []Rule{
+			{
+				Expr: "sumSeries(time('foo.bar',10))",
+				For:  promutils.NewDuration(10 * time.Millisecond),
+			},
+			{
+				Expr: "sum(up == 0 ) by (host)",
+			},
+		},
+	}, false, "either `record` or `alert` must be set")
+
+	// validate expressions
+	f(&Group{
+		Name: "test",
+		Rules: []Rule{
+			{
+				Record: "record",
+				Expr:   "up | 0",
+			},
+		},
+	}, true, "invalid expression")
+
+	f(&Group{
+		Name: "test thanos",
+		Type: NewRawType("thanos"),
+		Rules: []Rule{
+			{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+				"description": "{{ value|query }}",
+			}},
+		},
+	}, true, "unknown datasource type")
+
+	f(&Group{
+		Name: "test graphite",
+		Type: NewGraphiteType(),
+		Rules: []Rule{
+			{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+				"description": "some-description",
+			}},
+		},
+	}, true, "bad graphite expr")
+}
+
+func TestGroupValidate_Success(t *testing.T) {
+	f := func(group *Group, validateAnnotations, validateExpressions bool) {
+		t.Helper()
+
+		var validateTplFn ValidateTplFn
+		if validateAnnotations {
+			validateTplFn = notifier.ValidateTemplates
+		}
+		err := group.Validate(validateTplFn, validateExpressions)
+		if err != nil {
+			t.Fatalf("unexpected error: %s", err)
+		}
+	}
+
+	f(&Group{
+		Name: "test",
+		Rules: []Rule{
+			{
+				Record: "record",
+				Expr:   "up | 0",
+			},
+		},
+	}, false, false)
+
+	f(&Group{
+		Name: "test",
+		Rules: []Rule{
+			{
+				Alert: "alert",
+				Expr:  "up == 1",
+				Labels: map[string]string{
+					"summary": "{{ value|query }}",
 				},
 			},
-			expErr: "",
 		},
-		{
-			group: &Group{
-				Name: "test",
-				Rules: []Rule{
-					{
-						Record: "record",
-						Expr:   "up | 0",
-					},
-				},
-			},
-			expErr:              "invalid expression",
-			validateExpressions: true,
-		},
-		{
-			group: &Group{
-				Name: "test",
-				Rules: []Rule{
-					{
-						Alert: "alert",
-						Expr:  "up == 1",
-						Labels: map[string]string{
-							"summary": "{{ value|query }}",
-						},
-					},
-				},
-			},
-			expErr: "",
-		},
-		{
-			group: &Group{
-				Name: "test",
-				Rules: []Rule{
-					{
-						Alert: "alert",
-						Expr:  "up == 1",
-						Labels: map[string]string{
-							"summary": `
+	}, false, false)
+
+	// validate annotiations
+	f(&Group{
+		Name: "test",
+		Rules: []Rule{
+			{
+				Alert: "alert",
+				Expr:  "up == 1",
+				Labels: map[string]string{
+					"summary": `
 {{ with printf "node_memory_MemTotal{job='node',instance='%s'}" "localhost" | query }}
   {{ . | first | value | humanize1024 }}B
 {{ end }}`,
-						},
-					},
-				},
-			},
-			validateAnnotations: true,
-		},
-		{
-			group: &Group{
-				Name: "test",
-				Rules: []Rule{
-					{
-						Alert: "alert",
-						Expr:  "up == 1",
-					},
-					{
-						Alert: "alert",
-						Expr:  "up == 1",
-					},
-				},
-			},
-			expErr: "duplicate",
-		},
-		{
-			group: &Group{
-				Name: "test",
-				Rules: []Rule{
-					{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-						"summary": "{{ value|query }}",
-					}},
-					{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-						"summary": "{{ value|query }}",
-					}},
-				},
-			},
-			expErr: "duplicate",
-		},
-		{
-			group: &Group{
-				Name: "test",
-				Rules: []Rule{
-					{Record: "record", Expr: "up == 1", Labels: map[string]string{
-						"summary": "{{ value|query }}",
-					}},
-					{Record: "record", Expr: "up == 1", Labels: map[string]string{
-						"summary": "{{ value|query }}",
-					}},
-				},
-			},
-			expErr: "duplicate",
-		},
-		{
-			group: &Group{
-				Name: "test",
-				Rules: []Rule{
-					{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-						"summary": "{{ value|query }}",
-					}},
-					{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-						"description": "{{ value|query }}",
-					}},
-				},
-			},
-			expErr: "",
-		},
-		{
-			group: &Group{
-				Name: "test",
-				Rules: []Rule{
-					{Record: "alert", Expr: "up == 1", Labels: map[string]string{
-						"summary": "{{ value|query }}",
-					}},
-					{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-						"summary": "{{ value|query }}",
-					}},
-				},
-			},
-			expErr: "",
-		},
-		{
-			group: &Group{
-				Name: "test thanos",
-				Type: NewRawType("thanos"),
-				Rules: []Rule{
-					{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-						"description": "{{ value|query }}",
-					}},
-				},
-			},
-			validateExpressions: true,
-			expErr:              "unknown datasource type",
-		},
-		{
-			group: &Group{
-				Name: "test graphite",
-				Type: NewGraphiteType(),
-				Rules: []Rule{
-					{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-						"description": "some-description",
-					}},
-				},
-			},
-			validateExpressions: true,
-			expErr:              "",
-		},
-		{
-			group: &Group{
-				Name: "test prometheus",
-				Type: NewPrometheusType(),
-				Rules: []Rule{
-					{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-						"description": "{{ value|query }}",
-					}},
-				},
-			},
-			validateExpressions: true,
-			expErr:              "",
-		},
-		{
-			group: &Group{
-				Name: "test graphite inherit",
-				Type: NewGraphiteType(),
-				Rules: []Rule{
-					{
-						Expr: "sumSeries(time('foo.bar',10))",
-						For:  promutils.NewDuration(10 * time.Millisecond),
-					},
-					{
-						Expr: "sum(up == 0 ) by (host)",
-					},
 				},
 			},
 		},
-		{
-			group: &Group{
-				Name: "test graphite prometheus bad expr",
-				Type: NewGraphiteType(),
-				Rules: []Rule{
-					{
-						Expr: "sum(up == 0 ) by (host)",
-						For:  promutils.NewDuration(10 * time.Millisecond),
-					},
-					{
-						Expr: "sumSeries(time('foo.bar',10))",
-					},
-				},
-			},
-			expErr: "invalid rule",
-		},
-	}
+	}, true, false)
 
-	for _, tc := range testCases {
-		var validateTplFn ValidateTplFn
-		if tc.validateAnnotations {
-			validateTplFn = notifier.ValidateTemplates
-		}
-		err := tc.group.Validate(validateTplFn, tc.validateExpressions)
-		if err == nil {
-			if tc.expErr != "" {
-				t.Errorf("expected to get err %q; got nil insted", tc.expErr)
-			}
-			continue
-		}
-		if !strings.Contains(err.Error(), tc.expErr) {
-			t.Errorf("expected err to contain %q; got %q instead", tc.expErr, err)
-		}
-	}
+	// validate expressions
+	f(&Group{
+		Name: "test prometheus",
+		Type: NewPrometheusType(),
+		Rules: []Rule{
+			{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+				"description": "{{ value|query }}",
+			}},
+		},
+	}, false, true)
 }
 
-func TestHashRule(t *testing.T) {
-	testCases := []struct {
-		a, b  Rule
-		equal bool
-	}{
-		{
-			Rule{Record: "record", Expr: "up == 1"},
-			Rule{Record: "record", Expr: "up == 1"},
-			true,
-		},
-		{
-			Rule{Alert: "alert", Expr: "up == 1"},
-			Rule{Alert: "alert", Expr: "up == 1"},
-			true,
-		},
-		{
-			Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-				"foo": "bar",
-				"baz": "foo",
-			}},
-			Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-				"foo": "bar",
-				"baz": "foo",
-			}},
-			true,
-		},
-		{
-			Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-				"foo": "bar",
-				"baz": "foo",
-			}},
-			Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-				"baz": "foo",
-				"foo": "bar",
-			}},
-			true,
-		},
-		{
-			Rule{Alert: "record", Expr: "up == 1"},
-			Rule{Alert: "record", Expr: "up == 1"},
-			true,
-		},
-		{
-			Rule{Alert: "alert", Expr: "up == 1", For: promutils.NewDuration(time.Minute), KeepFiringFor: promutils.NewDuration(time.Minute)},
-			Rule{Alert: "alert", Expr: "up == 1"},
-			true,
-		},
-		{
-			Rule{Alert: "record", Expr: "up == 1"},
-			Rule{Record: "record", Expr: "up == 1"},
-			false,
-		},
-		{
-			Rule{Record: "record", Expr: "up == 1"},
-			Rule{Record: "record", Expr: "up == 2"},
-			false,
-		},
-		{
-			Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-				"foo": "bar",
-				"baz": "foo",
-			}},
-			Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-				"baz": "foo",
-				"foo": "baz",
-			}},
-			false,
-		},
-		{
-			Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-				"foo": "bar",
-				"baz": "foo",
-			}},
-			Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-				"baz": "foo",
-			}},
-			false,
-		},
-		{
-			Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-				"foo": "bar",
-				"baz": "foo",
-			}},
-			Rule{Alert: "alert", Expr: "up == 1"},
-			false,
-		},
-	}
-	for i, tc := range testCases {
-		aID, bID := HashRule(tc.a), HashRule(tc.b)
-		if tc.equal != (aID == bID) {
-			t.Fatalf("missmatch for rule %d", i)
+func TestHashRule_NotEqual(t *testing.T) {
+	f := func(a, b Rule) {
+		t.Helper()
+
+		aID, bID := HashRule(a), HashRule(b)
+		if aID == bID {
+			t.Fatalf("rule hashes mustn't be equal; got %d", aID)
 		}
 	}
+
+	f(Rule{Alert: "record", Expr: "up == 1"}, Rule{Record: "record", Expr: "up == 1"})
+
+	f(Rule{Record: "record", Expr: "up == 1"}, Rule{Record: "record", Expr: "up == 2"})
+
+	f(Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+		"foo": "bar",
+		"baz": "foo",
+	}}, Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+		"baz": "foo",
+		"foo": "baz",
+	}})
+
+	f(Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+		"foo": "bar",
+		"baz": "foo",
+	}}, Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+		"baz": "foo",
+	}})
+
+	f(Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+		"foo": "bar",
+		"baz": "foo",
+	}}, Rule{Alert: "alert", Expr: "up == 1"})
+}
+
+func TestHashRule_Equal(t *testing.T) {
+	f := func(a, b Rule) {
+		t.Helper()
+
+		aID, bID := HashRule(a), HashRule(b)
+		if aID != bID {
+			t.Fatalf("rule hashes must be equal; got %d and %d", aID, bID)
+		}
+	}
+
+	f(Rule{Record: "record", Expr: "up == 1"}, Rule{Record: "record", Expr: "up == 1"})
+
+	f(Rule{Alert: "alert", Expr: "up == 1"}, Rule{Alert: "alert", Expr: "up == 1"})
+
+	f(Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+		"foo": "bar",
+		"baz": "foo",
+	}}, Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+		"foo": "bar",
+		"baz": "foo",
+	}})
+
+	f(Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+		"foo": "bar",
+		"baz": "foo",
+	}}, Rule{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
+		"baz": "foo",
+		"foo": "bar",
+	}})
+
+	f(Rule{Alert: "record", Expr: "up == 1"}, Rule{Alert: "record", Expr: "up == 1"})
+
+	f(Rule{
+		Alert: "alert", Expr: "up == 1", For: promutils.NewDuration(time.Minute), KeepFiringFor: promutils.NewDuration(time.Minute),
+	}, Rule{Alert: "alert", Expr: "up == 1"})
 }
 
 func TestGroupChecksum(t *testing.T) {

app/vmalert/config/fs.go

@@ -32,14 +32,14 @@ var (
 	fsRegistry   = make(map[string]FS)
 )
 
-// readFromFS parses the given path list and inits FS for each item.
-// Once initialed, readFromFS will try to read and return files from each FS.
-// readFromFS returns an error if at least one FS failed to init.
+// ReadFromFS parses the given path list and inits FS for each item.
+// Once initialed, ReadFromFS will try to read and return files from each FS.
+// ReadFromFS returns an error if at least one FS failed to init.
 // The function can be called multiple times but each unique path
 // will be initialed only once.
 //
 // It is allowed to mix different FS types in path list.
-func readFromFS(paths []string) (map[string][]byte, error) {
+func ReadFromFS(paths []string) (map[string][]byte, error) {
 	var err error
 	result := make(map[string][]byte)
 	for _, path := range paths {

app/vmalert/config/log/logger.go

@@ -29,7 +29,7 @@ func (l *Logger) isDisabled() bool {
 }
 
 // Errorf logs error message.
-func (l *Logger) Errorf(format string, args ...interface{}) {
+func (l *Logger) Errorf(format string, args ...any) {
 	if l.isDisabled() {
 		return
 	}
@@ -37,7 +37,7 @@ func (l *Logger) Errorf(format string, args ...interface{}) {
 }
 
 // Warnf logs warning message.
-func (l *Logger) Warnf(format string, args ...interface{}) {
+func (l *Logger) Warnf(format string, args ...any) {
 	if l.isDisabled() {
 		return
 	}
@@ -45,7 +45,7 @@ func (l *Logger) Warnf(format string, args ...interface{}) {
 }
 
 // Infof logs info message.
-func (l *Logger) Infof(format string, args ...interface{}) {
+func (l *Logger) Infof(format string, args ...any) {
 	if l.isDisabled() {
 		return
 	}
@@ -54,6 +54,6 @@ func (l *Logger) Infof(format string, args ...interface{}) {
 
 // Panicf logs panic message and panics.
 // Panicf can't be suppressed
-func (l *Logger) Panicf(format string, args ...interface{}) {
+func (l *Logger) Panicf(format string, args ...any) {
 	logger.Panicf(format, args...)
 }

app/vmalert/config/log/logger_test.go

@@ -18,14 +18,14 @@ func TestOutput(t *testing.T) {
 
 	mustMatch := func(exp string) {
 		t.Helper()
+
 		if exp == "" {
 			if testOutput.String() != "" {
-				t.Errorf("expected output to be empty; got %q", testOutput.String())
-				return
+				t.Fatalf("expected output to be empty; got %q", testOutput.String())
 			}
 		}
 		if !strings.Contains(testOutput.String(), exp) {
-			t.Errorf("output %q should contain %q", testOutput.String(), exp)
+			t.Fatalf("output %q should contain %q", testOutput.String(), exp)
 		}
 		fmt.Println(testOutput.String())
 		testOutput.Reset()

app/vmalert/config/types.go

@@ -69,7 +69,7 @@ func (t *Type) ValidateExpr(expr string) error {
 }
 
 // UnmarshalYAML implements the yaml.Unmarshaler interface.
-func (t *Type) UnmarshalYAML(unmarshal func(interface{}) error) error {
+func (t *Type) UnmarshalYAML(unmarshal func(any) error) error {
 	var s string
 	if err := unmarshal(&s); err != nil {
 		return err
@@ -87,7 +87,7 @@ func (t *Type) UnmarshalYAML(unmarshal func(interface{}) error) error {
 }
 
 // MarshalYAML implements the yaml.Unmarshaler interface.
-func (t Type) MarshalYAML() (interface{}, error) {
+func (t Type) MarshalYAML() (any, error) {
 	return t.Name, nil
 }
 
@@ -98,7 +98,7 @@ type Header struct {
 }
 
 // UnmarshalYAML implements the yaml.Unmarshaler interface.
-func (h *Header) UnmarshalYAML(unmarshal func(interface{}) error) error {
+func (h *Header) UnmarshalYAML(unmarshal func(any) error) error {
 	var s string
 	if err := unmarshal(&s); err != nil {
 		return err

app/vmalert/datasource/init.go

@@ -12,6 +12,7 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputils"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/netutil"
 )
 
 var (
@@ -100,7 +101,7 @@ func Init(extraParams url.Values) (QuerierBuilder, error) {
 	if err != nil {
 		return nil, fmt.Errorf("failed to create transport: %w", err)
 	}
-	tr.DialContext = httputils.GetStatDialFunc("vmalert_datasource")
+	tr.DialContext = netutil.NewStatDialFunc("vmalert_datasource")
 	tr.DisableKeepAlives = *disableKeepAlive
 	tr.MaxIdleConnsPerHost = *maxIdleConnections
 	if tr.MaxIdleConns != 0 && tr.MaxIdleConns < tr.MaxIdleConnsPerHost {

app/vmalert/datasource/vm_prom_api.go

@@ -119,7 +119,7 @@ func (pi *promInstant) Unmarshal(b []byte) error {
 type promRange struct {
 	Result []struct {
 		Labels map[string]string `json:"metric"`
-		TVs    [][2]interface{}  `json:"values"`
+		TVs    [][2]any          `json:"values"`
 	} `json:"result"`
 }
 
@@ -147,7 +147,7 @@ func (r promRange) metrics() ([]Metric, error) {
 	return result, nil
 }
 
-type promScalar [2]interface{}
+type promScalar [2]any
 
 func (r promScalar) metrics() ([]Metric, error) {
 	var m Metric

app/vmalert/datasource/vm_test.go

@@ -31,26 +31,26 @@ var (
 func TestVMInstantQuery(t *testing.T) {
 	mux := http.NewServeMux()
 	mux.HandleFunc("/", func(_ http.ResponseWriter, _ *http.Request) {
-		t.Errorf("should not be called")
+		t.Fatalf("should not be called")
 	})
 	c := -1
 	mux.HandleFunc("/api/v1/query", func(w http.ResponseWriter, r *http.Request) {
 		c++
 		if r.Method != http.MethodPost {
-			t.Errorf("expected POST method got %s", r.Method)
+			t.Fatalf("expected POST method got %s", r.Method)
 		}
 		if name, pass, _ := r.BasicAuth(); name != basicAuthName || pass != basicAuthPass {
-			t.Errorf("expected %s:%s as basic auth got %s:%s", basicAuthName, basicAuthPass, name, pass)
+			t.Fatalf("expected %s:%s as basic auth got %s:%s", basicAuthName, basicAuthPass, name, pass)
 		}
 		if r.URL.Query().Get("query") != query {
-			t.Errorf("expected %s in query param, got %s", query, r.URL.Query().Get("query"))
+			t.Fatalf("expected %s in query param, got %s", query, r.URL.Query().Get("query"))
 		}
 		timeParam := r.URL.Query().Get("time")
 		if timeParam == "" {
-			t.Errorf("expected 'time' in query param, got nil instead")
+			t.Fatalf("expected 'time' in query param, got nil instead")
 		}
 		if _, err := time.Parse(time.RFC3339, timeParam); err != nil {
-			t.Errorf("failed to parse 'time' query param %q: %s", timeParam, err)
+			t.Fatalf("failed to parse 'time' query param %q: %s", timeParam, err)
 		}
 		switch c {
 		case 0:
@@ -197,13 +197,13 @@ func TestVMInstantQuery(t *testing.T) {
 func TestVMInstantQueryWithRetry(t *testing.T) {
 	mux := http.NewServeMux()
 	mux.HandleFunc("/", func(_ http.ResponseWriter, _ *http.Request) {
-		t.Errorf("should not be called")
+		t.Fatalf("should not be called")
 	})
 	c := -1
 	mux.HandleFunc("/api/v1/query", func(w http.ResponseWriter, r *http.Request) {
 		c++
 		if r.URL.Query().Get("query") != query {
-			t.Errorf("expected %s in query param, got %s", query, r.URL.Query().Get("query"))
+			t.Fatalf("expected %s in query param, got %s", query, r.URL.Query().Get("query"))
 		}
 		switch c {
 		case 0:
@@ -289,37 +289,37 @@ func metricsEqual(t *testing.T, gotM, expectedM []Metric) {
 func TestVMRangeQuery(t *testing.T) {
 	mux := http.NewServeMux()
 	mux.HandleFunc("/", func(_ http.ResponseWriter, _ *http.Request) {
-		t.Errorf("should not be called")
+		t.Fatalf("should not be called")
 	})
 	c := -1
 	mux.HandleFunc("/api/v1/query_range", func(w http.ResponseWriter, r *http.Request) {
 		c++
 		if r.Method != http.MethodPost {
-			t.Errorf("expected POST method got %s", r.Method)
+			t.Fatalf("expected POST method got %s", r.Method)
 		}
 		if name, pass, _ := r.BasicAuth(); name != basicAuthName || pass != basicAuthPass {
-			t.Errorf("expected %s:%s as basic auth got %s:%s", basicAuthName, basicAuthPass, name, pass)
+			t.Fatalf("expected %s:%s as basic auth got %s:%s", basicAuthName, basicAuthPass, name, pass)
 		}
 		if r.URL.Query().Get("query") != query {
-			t.Errorf("expected %s in query param, got %s", query, r.URL.Query().Get("query"))
+			t.Fatalf("expected %s in query param, got %s", query, r.URL.Query().Get("query"))
 		}
 		startTS := r.URL.Query().Get("start")
 		if startTS == "" {
-			t.Errorf("expected 'start' in query param, got nil instead")
+			t.Fatalf("expected 'start' in query param, got nil instead")
 		}
 		if _, err := time.Parse(time.RFC3339, startTS); err != nil {
-			t.Errorf("failed to parse 'start' query param: %s", err)
+			t.Fatalf("failed to parse 'start' query param: %s", err)
 		}
 		endTS := r.URL.Query().Get("end")
 		if endTS == "" {
-			t.Errorf("expected 'end' in query param, got nil instead")
+			t.Fatalf("expected 'end' in query param, got nil instead")
 		}
 		if _, err := time.Parse(time.RFC3339, endTS); err != nil {
-			t.Errorf("failed to parse 'end' query param: %s", err)
+			t.Fatalf("failed to parse 'end' query param: %s", err)
 		}
 		step := r.URL.Query().Get("step")
 		if step != "15s" {
-			t.Errorf("expected 'step' query param to be 15s; got %q instead", step)
+			t.Fatalf("expected 'step' query param to be 15s; got %q instead", step)
 		}
 		switch c {
 		case 0:
@@ -370,368 +370,299 @@ func TestVMRangeQuery(t *testing.T) {
 }
 
 func TestRequestParams(t *testing.T) {
-	authCfg, err := baCfg.NewConfig(".")
-	if err != nil {
-		t.Fatalf("unexpected: %s", err)
-	}
 	query := "up"
 	timestamp := time.Date(2001, 2, 3, 4, 5, 6, 0, time.UTC)
+
+	f := func(isQueryRange bool, vm *VMStorage, checkFn func(t *testing.T, r *http.Request)) {
+		t.Helper()
+
+		req, err := vm.newRequest(ctx)
+		if err != nil {
+			t.Fatalf("error in newRequest: %s", err)
+		}
+
+		switch vm.dataSourceType {
+		case "", datasourcePrometheus:
+			if isQueryRange {
+				vm.setPrometheusRangeReqParams(req, query, timestamp, timestamp)
+			} else {
+				vm.setPrometheusInstantReqParams(req, query, timestamp)
+			}
+		case datasourceGraphite:
+			vm.setGraphiteReqParams(req, query)
+		}
+
+		checkFn(t, req)
+	}
+
+	authCfg, err := baCfg.NewConfig(".")
+	if err != nil {
+		t.Fatalf("unexpected error: %s", err)
+	}
 	storage := VMStorage{
 		extraParams: url.Values{"round_digits": {"10"}},
 	}
-	testCases := []struct {
-		name       string
-		queryRange bool
-		vm         *VMStorage
-		checkFn    func(t *testing.T, r *http.Request)
-	}{
-		{
-			"prometheus path",
-			false,
-			&VMStorage{
-				dataSourceType: datasourcePrometheus,
-			},
-			func(t *testing.T, r *http.Request) {
-				checkEqualString(t, "/api/v1/query", r.URL.Path)
-			},
-		},
-		{
-			"prometheus prefix",
-			false,
-			&VMStorage{
-				dataSourceType:   datasourcePrometheus,
-				appendTypePrefix: true,
-			},
-			func(t *testing.T, r *http.Request) {
-				checkEqualString(t, "/prometheus/api/v1/query", r.URL.Path)
-			},
-		},
-		{
-			"prometheus range path",
-			true,
-			&VMStorage{
-				dataSourceType: datasourcePrometheus,
-			},
-			func(t *testing.T, r *http.Request) {
-				checkEqualString(t, "/api/v1/query_range", r.URL.Path)
-			},
-		},
-		{
-			"prometheus range prefix",
-			true,
-			&VMStorage{
-				dataSourceType:   datasourcePrometheus,
-				appendTypePrefix: true,
-			},
-			func(t *testing.T, r *http.Request) {
-				checkEqualString(t, "/prometheus/api/v1/query_range", r.URL.Path)
-			},
-		},
-		{
-			"graphite path",
-			false,
-			&VMStorage{
-				dataSourceType: datasourceGraphite,
-			},
-			func(t *testing.T, r *http.Request) {
-				checkEqualString(t, graphitePath, r.URL.Path)
-			},
-		},
-		{
-			"graphite prefix",
-			false,
-			&VMStorage{
-				dataSourceType:   datasourceGraphite,
-				appendTypePrefix: true,
-			},
-			func(t *testing.T, r *http.Request) {
-				checkEqualString(t, graphitePrefix+graphitePath, r.URL.Path)
-			},
-		},
-		{
-			"default params",
-			false,
-			&VMStorage{},
-			func(t *testing.T, r *http.Request) {
-				exp := url.Values{"query": {query}, "time": {timestamp.Format(time.RFC3339)}}
-				checkEqualString(t, exp.Encode(), r.URL.RawQuery)
-			},
-		},
-		{
-			"default range params",
-			true,
-			&VMStorage{},
-			func(t *testing.T, r *http.Request) {
-				ts := timestamp.Format(time.RFC3339)
-				exp := url.Values{"query": {query}, "start": {ts}, "end": {ts}}
-				checkEqualString(t, exp.Encode(), r.URL.RawQuery)
-			},
-		},
-		{
-			"basic auth",
-			false,
-			&VMStorage{authCfg: authCfg},
-			func(t *testing.T, r *http.Request) {
-				u, p, _ := r.BasicAuth()
-				checkEqualString(t, "foo", u)
-				checkEqualString(t, "bar", p)
-			},
-		},
-		{
-			"basic auth range",
-			true,
-			&VMStorage{authCfg: authCfg},
-			func(t *testing.T, r *http.Request) {
-				u, p, _ := r.BasicAuth()
-				checkEqualString(t, "foo", u)
-				checkEqualString(t, "bar", p)
-			},
-		},
-		{
-			"evaluation interval",
-			false,
-			&VMStorage{
-				evaluationInterval: 15 * time.Second,
-			},
-			func(t *testing.T, r *http.Request) {
-				evalInterval := 15 * time.Second
-				exp := url.Values{"query": {query}, "step": {evalInterval.String()}, "time": {timestamp.Format(time.RFC3339)}}
-				checkEqualString(t, exp.Encode(), r.URL.RawQuery)
-			},
-		},
-		{
-			"step override",
-			false,
-			&VMStorage{
-				queryStep: time.Minute,
-			},
-			func(t *testing.T, r *http.Request) {
-				exp := url.Values{
-					"query": {query},
-					"step":  {fmt.Sprintf("%ds", int(time.Minute.Seconds()))},
-					"time":  {timestamp.Format(time.RFC3339)},
-				}
-				checkEqualString(t, exp.Encode(), r.URL.RawQuery)
-			},
-		},
-		{
-			"step to seconds",
-			false,
-			&VMStorage{
-				evaluationInterval: 3 * time.Hour,
-			},
-			func(t *testing.T, r *http.Request) {
-				evalInterval := 3 * time.Hour
-				exp := url.Values{"query": {query}, "step": {fmt.Sprintf("%ds", int(evalInterval.Seconds()))}, "time": {timestamp.Format(time.RFC3339)}}
-				checkEqualString(t, exp.Encode(), r.URL.RawQuery)
-			},
-		},
-		{
-			"prometheus extra params",
-			false,
-			&VMStorage{
-				extraParams: url.Values{"round_digits": {"10"}},
-			},
-			func(t *testing.T, r *http.Request) {
-				exp := url.Values{"query": {query}, "round_digits": {"10"}, "time": {timestamp.Format(time.RFC3339)}}
-				checkEqualString(t, exp.Encode(), r.URL.RawQuery)
-			},
-		},
-		{
-			"prometheus extra params range",
-			true,
-			&VMStorage{
-				extraParams: url.Values{
-					"nocache":      {"1"},
-					"max_lookback": {"1h"},
-				},
-			},
-			func(t *testing.T, r *http.Request) {
-				exp := url.Values{
-					"query":        {query},
-					"end":          {timestamp.Format(time.RFC3339)},
-					"start":        {timestamp.Format(time.RFC3339)},
-					"nocache":      {"1"},
-					"max_lookback": {"1h"},
-				}
-				checkEqualString(t, exp.Encode(), r.URL.RawQuery)
-			},
-		},
-		{
-			"custom params overrides the original params",
-			false,
-			storage.Clone().ApplyParams(QuerierParams{
-				QueryParams: url.Values{"round_digits": {"2"}},
-			}),
-			func(t *testing.T, r *http.Request) {
-				exp := url.Values{"query": {query}, "round_digits": {"2"}, "time": {timestamp.Format(time.RFC3339)}}
-				checkEqualString(t, exp.Encode(), r.URL.RawQuery)
-			},
-		},
-		{
-			"allow duplicates in query params",
-			false,
-			storage.Clone().ApplyParams(QuerierParams{
-				QueryParams: url.Values{"extra_labels": {"env=dev", "foo=bar"}},
-			}),
-			func(t *testing.T, r *http.Request) {
-				exp := url.Values{"query": {query}, "round_digits": {"10"}, "extra_labels": {"env=dev", "foo=bar"}, "time": {timestamp.Format(time.RFC3339)}}
-				checkEqualString(t, exp.Encode(), r.URL.RawQuery)
-			},
-		},
-		{
-			"graphite extra params",
-			false,
-			&VMStorage{
-				dataSourceType: datasourceGraphite,
-				extraParams: url.Values{
-					"nocache":      {"1"},
-					"max_lookback": {"1h"},
-				},
-			},
-			func(t *testing.T, r *http.Request) {
-				exp := fmt.Sprintf("format=json&from=-5min&max_lookback=1h&nocache=1&target=%s&until=now", query)
-				checkEqualString(t, exp, r.URL.RawQuery)
-			},
-		},
-		{
-			"graphite extra params allows to override from",
-			false,
-			&VMStorage{
-				dataSourceType: datasourceGraphite,
-				extraParams: url.Values{
-					"from": {"-10m"},
-				},
-			},
-			func(t *testing.T, r *http.Request) {
-				exp := fmt.Sprintf("format=json&from=-10m&target=%s&until=now", query)
-				checkEqualString(t, exp, r.URL.RawQuery)
-			},
-		},
-	}
 
-	for _, tc := range testCases {
-		t.Run(tc.name, func(t *testing.T) {
-			req, err := tc.vm.newRequest(ctx)
-			if err != nil {
-				t.Fatal(err)
-			}
-			switch tc.vm.dataSourceType {
-			case "", datasourcePrometheus:
-				if tc.queryRange {
-					tc.vm.setPrometheusRangeReqParams(req, query, timestamp, timestamp)
-				} else {
-					tc.vm.setPrometheusInstantReqParams(req, query, timestamp)
-				}
-			case datasourceGraphite:
-				tc.vm.setGraphiteReqParams(req, query)
-			}
-			tc.checkFn(t, req)
-		})
-	}
+	// prometheus path
+	f(false, &VMStorage{
+		dataSourceType: datasourcePrometheus,
+	}, func(t *testing.T, r *http.Request) {
+		checkEqualString(t, "/api/v1/query", r.URL.Path)
+	})
+
+	// prometheus prefix
+	f(false, &VMStorage{
+		dataSourceType:   datasourcePrometheus,
+		appendTypePrefix: true,
+	}, func(t *testing.T, r *http.Request) {
+		checkEqualString(t, "/prometheus/api/v1/query", r.URL.Path)
+	})
+
+	// prometheus range path
+	f(true, &VMStorage{
+		dataSourceType: datasourcePrometheus,
+	}, func(t *testing.T, r *http.Request) {
+		checkEqualString(t, "/api/v1/query_range", r.URL.Path)
+	})
+
+	// prometheus range prefix
+	f(true, &VMStorage{
+		dataSourceType:   datasourcePrometheus,
+		appendTypePrefix: true,
+	}, func(t *testing.T, r *http.Request) {
+		checkEqualString(t, "/prometheus/api/v1/query_range", r.URL.Path)
+	})
+
+	// graphite path
+	f(false, &VMStorage{
+		dataSourceType: datasourceGraphite,
+	}, func(t *testing.T, r *http.Request) {
+		checkEqualString(t, graphitePath, r.URL.Path)
+	})
+
+	// graphite prefix
+	f(false, &VMStorage{
+		dataSourceType:   datasourceGraphite,
+		appendTypePrefix: true,
+	}, func(t *testing.T, r *http.Request) {
+		checkEqualString(t, graphitePrefix+graphitePath, r.URL.Path)
+	})
+
+	// default params
+	f(false, &VMStorage{}, func(t *testing.T, r *http.Request) {
+		exp := url.Values{"query": {query}, "time": {timestamp.Format(time.RFC3339)}}
+		checkEqualString(t, exp.Encode(), r.URL.RawQuery)
+	})
+
+	// default range params
+	f(true, &VMStorage{}, func(t *testing.T, r *http.Request) {
+		ts := timestamp.Format(time.RFC3339)
+		exp := url.Values{"query": {query}, "start": {ts}, "end": {ts}}
+		checkEqualString(t, exp.Encode(), r.URL.RawQuery)
+	})
+
+	// basic auth
+	f(false, &VMStorage{
+		authCfg: authCfg,
+	}, func(t *testing.T, r *http.Request) {
+		u, p, _ := r.BasicAuth()
+		checkEqualString(t, "foo", u)
+		checkEqualString(t, "bar", p)
+	})
+
+	// basic auth range
+	f(true, &VMStorage{
+		authCfg: authCfg,
+	}, func(t *testing.T, r *http.Request) {
+		u, p, _ := r.BasicAuth()
+		checkEqualString(t, "foo", u)
+		checkEqualString(t, "bar", p)
+	})
+
+	// evaluation interval
+	f(false, &VMStorage{
+		evaluationInterval: 15 * time.Second,
+	}, func(t *testing.T, r *http.Request) {
+		evalInterval := 15 * time.Second
+		exp := url.Values{"query": {query}, "step": {evalInterval.String()}, "time": {timestamp.Format(time.RFC3339)}}
+		checkEqualString(t, exp.Encode(), r.URL.RawQuery)
+	})
+
+	// step override
+	f(false, &VMStorage{
+		queryStep: time.Minute,
+	}, func(t *testing.T, r *http.Request) {
+		exp := url.Values{
+			"query": {query},
+			"step":  {fmt.Sprintf("%ds", int(time.Minute.Seconds()))},
+			"time":  {timestamp.Format(time.RFC3339)},
+		}
+		checkEqualString(t, exp.Encode(), r.URL.RawQuery)
+	})
+
+	//  step to seconds
+	f(false, &VMStorage{
+		evaluationInterval: 3 * time.Hour,
+	}, func(t *testing.T, r *http.Request) {
+		evalInterval := 3 * time.Hour
+		exp := url.Values{"query": {query}, "step": {fmt.Sprintf("%ds", int(evalInterval.Seconds()))}, "time": {timestamp.Format(time.RFC3339)}}
+		checkEqualString(t, exp.Encode(), r.URL.RawQuery)
+	})
+
+	// prometheus extra params
+	f(false, &VMStorage{
+		extraParams: url.Values{"round_digits": {"10"}},
+	}, func(t *testing.T, r *http.Request) {
+		exp := url.Values{"query": {query}, "round_digits": {"10"}, "time": {timestamp.Format(time.RFC3339)}}
+		checkEqualString(t, exp.Encode(), r.URL.RawQuery)
+	})
+
+	// prometheus extra params range
+	f(true, &VMStorage{
+		extraParams: url.Values{
+			"nocache":      {"1"},
+			"max_lookback": {"1h"},
+		},
+	}, func(t *testing.T, r *http.Request) {
+		exp := url.Values{
+			"query":        {query},
+			"end":          {timestamp.Format(time.RFC3339)},
+			"start":        {timestamp.Format(time.RFC3339)},
+			"nocache":      {"1"},
+			"max_lookback": {"1h"},
+		}
+		checkEqualString(t, exp.Encode(), r.URL.RawQuery)
+	})
+
+	// custom params overrides the original params
+	f(false, storage.Clone().ApplyParams(QuerierParams{
+		QueryParams: url.Values{"round_digits": {"2"}},
+	}), func(t *testing.T, r *http.Request) {
+		exp := url.Values{"query": {query}, "round_digits": {"2"}, "time": {timestamp.Format(time.RFC3339)}}
+		checkEqualString(t, exp.Encode(), r.URL.RawQuery)
+	})
+
+	// allow duplicates in query params
+	f(false, storage.Clone().ApplyParams(QuerierParams{
+		QueryParams: url.Values{"extra_labels": {"env=dev", "foo=bar"}},
+	}), func(t *testing.T, r *http.Request) {
+		exp := url.Values{"query": {query}, "round_digits": {"10"}, "extra_labels": {"env=dev", "foo=bar"}, "time": {timestamp.Format(time.RFC3339)}}
+		checkEqualString(t, exp.Encode(), r.URL.RawQuery)
+	})
+
+	// graphite extra params
+	f(false, &VMStorage{
+		dataSourceType: datasourceGraphite,
+		extraParams: url.Values{
+			"nocache":      {"1"},
+			"max_lookback": {"1h"},
+		},
+	}, func(t *testing.T, r *http.Request) {
+		exp := fmt.Sprintf("format=json&from=-5min&max_lookback=1h&nocache=1&target=%s&until=now", query)
+		checkEqualString(t, exp, r.URL.RawQuery)
+	})
+
+	// graphite extra params allows to override from
+	f(false, &VMStorage{
+		dataSourceType: datasourceGraphite,
+		extraParams: url.Values{
+			"from": {"-10m"},
+		},
+	}, func(t *testing.T, r *http.Request) {
+		exp := fmt.Sprintf("format=json&from=-10m&target=%s&until=now", query)
+		checkEqualString(t, exp, r.URL.RawQuery)
+	})
 }
 
 func TestHeaders(t *testing.T) {
-	testCases := []struct {
-		name    string
-		vmFn    func() *VMStorage
-		checkFn func(t *testing.T, r *http.Request)
-	}{
-		{
-			name: "basic auth",
-			vmFn: func() *VMStorage {
-				cfg, err := utils.AuthConfig(utils.WithBasicAuth("foo", "bar", ""))
-				if err != nil {
-					t.Errorf("Error get auth config: %s", err)
-				}
-				return &VMStorage{authCfg: cfg}
-			},
-			checkFn: func(t *testing.T, r *http.Request) {
-				u, p, _ := r.BasicAuth()
-				checkEqualString(t, "foo", u)
-				checkEqualString(t, "bar", p)
-			},
-		},
-		{
-			name: "bearer auth",
-			vmFn: func() *VMStorage {
-				cfg, err := utils.AuthConfig(utils.WithBearer("foo", ""))
-				if err != nil {
-					t.Errorf("Error get auth config: %s", err)
-				}
-				return &VMStorage{authCfg: cfg}
-			},
-			checkFn: func(t *testing.T, r *http.Request) {
-				reqToken := r.Header.Get("Authorization")
-				splitToken := strings.Split(reqToken, "Bearer ")
-				if len(splitToken) != 2 {
-					t.Errorf("expected two items got %d", len(splitToken))
-				}
-				token := splitToken[1]
-				checkEqualString(t, "foo", token)
-			},
-		},
-		{
-			name: "custom extraHeaders",
-			vmFn: func() *VMStorage {
-				return &VMStorage{extraHeaders: []keyValue{
-					{key: "Foo", value: "bar"},
-					{key: "Baz", value: "qux"},
-				}}
-			},
-			checkFn: func(t *testing.T, r *http.Request) {
-				h1 := r.Header.Get("Foo")
-				checkEqualString(t, "bar", h1)
-				h2 := r.Header.Get("Baz")
-				checkEqualString(t, "qux", h2)
-			},
-		},
-		{
-			name: "custom header overrides basic auth",
-			vmFn: func() *VMStorage {
-				cfg, err := utils.AuthConfig(utils.WithBasicAuth("foo", "bar", ""))
-				if err != nil {
-					t.Errorf("Error get auth config: %s", err)
-				}
-				return &VMStorage{
-					authCfg: cfg,
-					extraHeaders: []keyValue{
-						{key: "Authorization", value: "Basic QWxhZGRpbjpvcGVuIHNlc2FtZQ=="},
-					},
-				}
-			},
-			checkFn: func(t *testing.T, r *http.Request) {
-				u, p, _ := r.BasicAuth()
-				checkEqualString(t, "Aladdin", u)
-				checkEqualString(t, "open sesame", p)
-			},
-		},
-	}
-	for _, tt := range testCases {
-		t.Run(tt.name, func(t *testing.T) {
-			vm := tt.vmFn()
-			req, err := vm.newQueryRequest(ctx, "foo", time.Now())
-			if err != nil {
-				t.Fatal(err)
-			}
-			tt.checkFn(t, req)
-		})
+	f := func(vmFn func() *VMStorage, checkFn func(t *testing.T, r *http.Request)) {
+		t.Helper()
+
+		vm := vmFn()
+		req, err := vm.newQueryRequest(ctx, "foo", time.Now())
+		if err != nil {
+			t.Fatalf("error in newQueryRequest: %s", err)
+		}
+		checkFn(t, req)
 	}
+
+	// basic auth
+	f(func() *VMStorage {
+		cfg, err := utils.AuthConfig(utils.WithBasicAuth("foo", "bar", ""))
+		if err != nil {
+			t.Fatalf("Error get auth config: %s", err)
+		}
+		return &VMStorage{authCfg: cfg}
+	}, func(t *testing.T, r *http.Request) {
+		u, p, _ := r.BasicAuth()
+		checkEqualString(t, "foo", u)
+		checkEqualString(t, "bar", p)
+	})
+
+	// bearer auth
+	f(func() *VMStorage {
+		cfg, err := utils.AuthConfig(utils.WithBearer("foo", ""))
+		if err != nil {
+			t.Fatalf("Error get auth config: %s", err)
+		}
+		return &VMStorage{authCfg: cfg}
+	}, func(t *testing.T, r *http.Request) {
+		reqToken := r.Header.Get("Authorization")
+		splitToken := strings.Split(reqToken, "Bearer ")
+		if len(splitToken) != 2 {
+			t.Fatalf("expected two items got %d", len(splitToken))
+		}
+		token := splitToken[1]
+		checkEqualString(t, "foo", token)
+	})
+
+	// custom extraHeaders
+	f(func() *VMStorage {
+		return &VMStorage{extraHeaders: []keyValue{
+			{key: "Foo", value: "bar"},
+			{key: "Baz", value: "qux"},
+		}}
+	}, func(t *testing.T, r *http.Request) {
+		h1 := r.Header.Get("Foo")
+		checkEqualString(t, "bar", h1)
+		h2 := r.Header.Get("Baz")
+		checkEqualString(t, "qux", h2)
+	})
+
+	// custom header overrides basic auth
+	f(func() *VMStorage {
+		cfg, err := utils.AuthConfig(utils.WithBasicAuth("foo", "bar", ""))
+		if err != nil {
+			t.Fatalf("Error get auth config: %s", err)
+		}
+		return &VMStorage{
+			authCfg: cfg,
+			extraHeaders: []keyValue{
+				{key: "Authorization", value: "Basic QWxhZGRpbjpvcGVuIHNlc2FtZQ=="},
+			},
+		}
+	}, func(t *testing.T, r *http.Request) {
+		u, p, _ := r.BasicAuth()
+		checkEqualString(t, "Aladdin", u)
+		checkEqualString(t, "open sesame", p)
+	})
 }
 
 func checkEqualString(t *testing.T, exp, got string) {
 	t.Helper()
+
 	if got != exp {
-		t.Errorf("expected to get: \n%q; \ngot: \n%q", exp, got)
+		t.Fatalf("expected to get: \n%q; \ngot: \n%q", exp, got)
 	}
 }
 
 func expectError(t *testing.T, err error, exp string) {
 	t.Helper()
+
 	if err == nil {
-		t.Errorf("expected non-nil error")
+		t.Fatalf("expected non-nil error")
 	}
 	if !strings.Contains(err.Error(), exp) {
-		t.Errorf("expected error %q to contain %q", err, exp)
+		t.Fatalf("expected error %q to contain %q", err, exp)
 	}
 }

app/vmalert/main.go

@@ -72,7 +72,7 @@ absolute path to all .tpl files in root.
 	externalAlertSource = flag.String("external.alert.source", "", `External Alert Source allows to override the Source link for alerts sent to AlertManager `+
 		`for cases where you want to build a custom link to Grafana, Prometheus or any other service. `+
 		`Supports templating - see https://docs.victoriametrics.com/vmalert/#templating . `+
-		`For example, link to Grafana: -external.alert.source='explore?orgId=1&left={"datasource":"VictoriaMetrics","queries":[{"expr":{{$expr|jsonEscape|queryEscape}},"refId":"A"}],"range":{"from":"now-1h","to":"now"}}'. `+
+		`For example, link to Grafana: -external.alert.source='explore?orgId=1&left={"datasource":"VictoriaMetrics","queries":[{"expr":{{.Expr|jsonEscape|queryEscape}},"refId":"A"}],"range":{"from":"now-1h","to":"now"}}'. `+
 		`Link to VMUI: -external.alert.source='vmui/#/?g0.expr={{.Expr|queryEscape}}'. `+
 		`If empty 'vmalert/alert?group_id={{.GroupID}}&alert_id={{.AlertID}}' is used.`)
 	externalLabels = flagutil.NewArrayString("external.label", "Optional label in the form 'Name=value' to add to all generated recording rules and alerts. "+
@@ -132,7 +132,7 @@ func main() {
 		validateTplFn = notifier.ValidateTemplates
 	}
 
-	if *replayFrom != "" || *replayTo != "" {
+	if *replayFrom != "" {
 		rw, err := remotewrite.Init(context.Background())
 		if err != nil {
 			logger.Fatalf("failed to init remoteWrite: %s", err)
@@ -153,6 +153,7 @@ func main() {
 		if err := replay(groupsCfg, q, rw); err != nil {
 			logger.Fatalf("replay failed: %s", err)
 		}
+		logger.Infof("replay succeed!")
 		return
 	}
 

app/vmalert/main_test.go

@@ -25,26 +25,26 @@ func TestGetExternalURL(t *testing.T) {
 	invalidURL := "victoriametrics.com/path"
 	_, err := getExternalURL(invalidURL)
 	if err == nil {
-		t.Errorf("expected error, got nil")
+		t.Fatalf("expected error, got nil")
 	}
 
 	expURL := "https://victoriametrics.com/path"
 	u, err := getExternalURL(expURL)
 	if err != nil {
-		t.Errorf("unexpected error %s", err)
+		t.Fatalf("unexpected error %s", err)
 	}
 	if u.String() != expURL {
-		t.Errorf("unexpected url: want %q, got %s", expURL, u.String())
+		t.Fatalf("unexpected url: want %q, got %s", expURL, u.String())
 	}
 
 	h, _ := os.Hostname()
 	expURL = fmt.Sprintf("http://%s:8880", h)
 	u, err = getExternalURL("")
 	if err != nil {
-		t.Errorf("unexpected error %s", err)
+		t.Fatalf("unexpected error %s", err)
 	}
 	if u.String() != expURL {
-		t.Errorf("unexpected url: want %s, got %s", expURL, u.String())
+		t.Fatalf("unexpected url: want %s, got %s", expURL, u.String())
 	}
 }
 
@@ -53,22 +53,22 @@ func TestGetAlertURLGenerator(t *testing.T) {
 	u, _ := url.Parse("https://victoriametrics.com/path")
 	fn, err := getAlertURLGenerator(u, "", false)
 	if err != nil {
-		t.Errorf("unexpected error %s", err)
+		t.Fatalf("unexpected error %s", err)
 	}
 	exp := fmt.Sprintf("https://victoriametrics.com/path/vmalert/alert?%s=42&%s=2", paramGroupID, paramAlertID)
 	if exp != fn(testAlert) {
-		t.Errorf("unexpected url want %s, got %s", exp, fn(testAlert))
+		t.Fatalf("unexpected url want %s, got %s", exp, fn(testAlert))
 	}
 	_, err = getAlertURLGenerator(nil, "foo?{{invalid}}", true)
 	if err == nil {
-		t.Errorf("expected template validation error got nil")
+		t.Fatalf("expected template validation error got nil")
 	}
 	fn, err = getAlertURLGenerator(u, "foo?query={{$value}}&ds={{ $labels.tenant }}", true)
 	if err != nil {
-		t.Errorf("unexpected error %s", err)
+		t.Fatalf("unexpected error %s", err)
 	}
 	if exp := "https://victoriametrics.com/path/foo?query=4&ds=baz"; exp != fn(testAlert) {
-		t.Errorf("unexpected url want %s, got %s", exp, fn(testAlert))
+		t.Fatalf("unexpected url want %s, got %s", exp, fn(testAlert))
 	}
 }
 

app/vmalert/manager_test.go

@@ -82,9 +82,8 @@ func TestManagerUpdateConcurrent(t *testing.T) {
 	wg.Wait()
 }
 
-// TestManagerUpdate tests sequential configuration
-// updates.
-func TestManagerUpdate(t *testing.T) {
+// TestManagerUpdate tests sequential configuration updates.
+func TestManagerUpdate_Success(t *testing.T) {
 	const defaultEvalInterval = time.Second * 30
 	currentEvalInterval := *evaluationInterval
 	*evaluationInterval = defaultEvalInterval
@@ -120,145 +119,127 @@ func TestManagerUpdate(t *testing.T) {
 		}
 	)
 
-	testCases := []struct {
-		name       string
-		initPath   string
-		updatePath string
-		want       []*rule.Group
-	}{
-		{
-			name:       "update good rules",
-			initPath:   "config/testdata/rules/rules0-good.rules",
-			updatePath: "config/testdata/dir/rules1-good.rules",
-			want: []*rule.Group{
-				{
-					File:     "config/testdata/dir/rules1-good.rules",
-					Name:     "duplicatedGroupDiffFiles",
-					Type:     config.NewPrometheusType(),
-					Interval: defaultEvalInterval,
-					Rules: []rule.Rule{
-						&rule.AlertingRule{
-							Name:   "VMRows",
-							Expr:   "vm_rows > 0",
-							For:    5 * time.Minute,
-							Labels: map[string]string{"dc": "gcp", "label": "bar"},
-							Annotations: map[string]string{
-								"summary":     "{{ $value }}",
-								"description": "{{$labels}}",
-							},
-						},
-					},
-				},
-			},
-		},
-		{
-			name:       "update good rules from 1 to 2 groups",
-			initPath:   "config/testdata/dir/rules/rules1-good.rules",
-			updatePath: "config/testdata/rules/rules0-good.rules",
-			want: []*rule.Group{
-				{
-					File:     "config/testdata/rules/rules0-good.rules",
-					Name:     "groupGorSingleAlert",
-					Type:     config.NewPrometheusType(),
-					Interval: defaultEvalInterval,
-					Rules:    []rule.Rule{VMRows},
-				},
-				{
-					File:     "config/testdata/rules/rules0-good.rules",
-					Interval: defaultEvalInterval,
-					Type:     config.NewPrometheusType(),
-					Name:     "TestGroup",
-					Rules: []rule.Rule{
-						Conns,
-						ExampleAlertAlwaysFiring,
-					},
-				},
-			},
-		},
-		{
-			name:       "update with one bad rule file",
-			initPath:   "config/testdata/rules/rules0-good.rules",
-			updatePath: "config/testdata/dir/rules2-bad.rules",
-			want: []*rule.Group{
-				{
-					File:     "config/testdata/rules/rules0-good.rules",
-					Name:     "groupGorSingleAlert",
-					Type:     config.NewPrometheusType(),
-					Interval: defaultEvalInterval,
-					Rules:    []rule.Rule{VMRows},
-				},
-				{
-					File:     "config/testdata/rules/rules0-good.rules",
-					Interval: defaultEvalInterval,
-					Name:     "TestGroup",
-					Type:     config.NewPrometheusType(),
-					Rules: []rule.Rule{
-						Conns,
-						ExampleAlertAlwaysFiring,
-					},
-				},
-			},
-		},
-		{
-			name:       "update empty dir rules from 0 to 2 groups",
-			initPath:   "config/testdata/empty/*",
-			updatePath: "config/testdata/rules/rules0-good.rules",
-			want: []*rule.Group{
-				{
-					File:     "config/testdata/rules/rules0-good.rules",
-					Name:     "groupGorSingleAlert",
-					Type:     config.NewPrometheusType(),
-					Interval: defaultEvalInterval,
-					Rules:    []rule.Rule{VMRows},
-				},
-				{
-					File:     "config/testdata/rules/rules0-good.rules",
-					Interval: defaultEvalInterval,
-					Type:     config.NewPrometheusType(),
-					Name:     "TestGroup",
-					Rules: []rule.Rule{
-						Conns,
-						ExampleAlertAlwaysFiring,
-					},
-				},
-			},
-		},
+	f := func(initPath, updatePath string, groupsExpected []*rule.Group) {
+		t.Helper()
+
+		ctx, cancel := context.WithCancel(context.TODO())
+		m := &manager{
+			groups:         make(map[uint64]*rule.Group),
+			querierBuilder: &datasource.FakeQuerier{},
+			notifiers:      func() []notifier.Notifier { return []notifier.Notifier{&notifier.FakeNotifier{}} },
+		}
+
+		cfgInit := loadCfg(t, []string{initPath}, true, true)
+		if err := m.update(ctx, cfgInit, false); err != nil {
+			t.Fatalf("failed to complete initial rules update: %s", err)
+		}
+
+		cfgUpdate, err := config.Parse([]string{updatePath}, notifier.ValidateTemplates, true)
+		if err == nil { // update can fail and that's expected
+			_ = m.update(ctx, cfgUpdate, false)
+		}
+		if len(groupsExpected) != len(m.groups) {
+			t.Fatalf("unexpected number of groups; got %d; want %d", len(m.groups), len(groupsExpected))
+		}
+
+		for _, wantG := range groupsExpected {
+			gotG, ok := m.groups[wantG.ID()]
+			if !ok {
+				t.Fatalf("expected to have group %q", wantG.Name)
+			}
+			compareGroups(t, wantG, gotG)
+		}
+
+		cancel()
+		m.close()
 	}
-	for _, tc := range testCases {
-		t.Run(tc.name, func(t *testing.T) {
-			ctx, cancel := context.WithCancel(context.TODO())
-			m := &manager{
-				groups:         make(map[uint64]*rule.Group),
-				querierBuilder: &datasource.FakeQuerier{},
-				notifiers:      func() []notifier.Notifier { return []notifier.Notifier{&notifier.FakeNotifier{}} },
-			}
 
-			cfgInit := loadCfg(t, []string{tc.initPath}, true, true)
-			if err := m.update(ctx, cfgInit, false); err != nil {
-				t.Fatalf("failed to complete initial rules update: %s", err)
-			}
+	// update good rules
+	f("config/testdata/rules/rules0-good.rules", "config/testdata/dir/rules1-good.rules", []*rule.Group{
+		{
+			File:     "config/testdata/dir/rules1-good.rules",
+			Name:     "duplicatedGroupDiffFiles",
+			Type:     config.NewPrometheusType(),
+			Interval: defaultEvalInterval,
+			Rules: []rule.Rule{
+				&rule.AlertingRule{
+					Name:   "VMRows",
+					Expr:   "vm_rows > 0",
+					For:    5 * time.Minute,
+					Labels: map[string]string{"dc": "gcp", "label": "bar"},
+					Annotations: map[string]string{
+						"summary":     "{{ $value }}",
+						"description": "{{$labels}}",
+					},
+				},
+			},
+		},
+	})
 
-			cfgUpdate, err := config.Parse([]string{tc.updatePath}, notifier.ValidateTemplates, true)
-			if err == nil { // update can fail and that's expected
-				_ = m.update(ctx, cfgUpdate, false)
-			}
-			if len(tc.want) != len(m.groups) {
-				t.Fatalf("\nwant number of groups: %d;\ngot: %d ", len(tc.want), len(m.groups))
-			}
+	// update good rules from 1 to 2 groups
+	f("config/testdata/dir/rules/rules1-good.rules", "config/testdata/rules/rules0-good.rules", []*rule.Group{
+		{
+			File:     "config/testdata/rules/rules0-good.rules",
+			Name:     "groupGorSingleAlert",
+			Type:     config.NewPrometheusType(),
+			Interval: defaultEvalInterval,
+			Rules:    []rule.Rule{VMRows},
+		},
+		{
+			File:     "config/testdata/rules/rules0-good.rules",
+			Interval: defaultEvalInterval,
+			Type:     config.NewPrometheusType(),
+			Name:     "TestGroup",
+			Rules: []rule.Rule{
+				Conns,
+				ExampleAlertAlwaysFiring,
+			},
+		},
+	})
 
-			for _, wantG := range tc.want {
-				gotG, ok := m.groups[wantG.ID()]
-				if !ok {
-					t.Fatalf("expected to have group %q", wantG.Name)
-				}
-				compareGroups(t, wantG, gotG)
-			}
+	// update with one bad rule file
+	f("config/testdata/rules/rules0-good.rules", "config/testdata/dir/rules2-bad.rules", []*rule.Group{
+		{
+			File:     "config/testdata/rules/rules0-good.rules",
+			Name:     "groupGorSingleAlert",
+			Type:     config.NewPrometheusType(),
+			Interval: defaultEvalInterval,
+			Rules:    []rule.Rule{VMRows},
+		},
+		{
+			File:     "config/testdata/rules/rules0-good.rules",
+			Interval: defaultEvalInterval,
+			Name:     "TestGroup",
+			Type:     config.NewPrometheusType(),
+			Rules: []rule.Rule{
+				Conns,
+				ExampleAlertAlwaysFiring,
+			},
+		},
+	})
 
-			cancel()
-			m.close()
-		})
-	}
+	// update empty dir rules from 0 to 2 groups
+	f("config/testdata/empty/*", "config/testdata/rules/rules0-good.rules", []*rule.Group{
+		{
+			File:     "config/testdata/rules/rules0-good.rules",
+			Name:     "groupGorSingleAlert",
+			Type:     config.NewPrometheusType(),
+			Interval: defaultEvalInterval,
+			Rules:    []rule.Rule{VMRows},
+		},
+		{
+			File:     "config/testdata/rules/rules0-good.rules",
+			Interval: defaultEvalInterval,
+			Type:     config.NewPrometheusType(),
+			Name:     "TestGroup",
+			Rules: []rule.Rule{
+				Conns,
+				ExampleAlertAlwaysFiring,
+			},
+		},
+	})
 }
+
 func compareGroups(t *testing.T, a, b *rule.Group) {
 	t.Helper()
 	if a.Name != b.Name {
@@ -285,82 +266,59 @@ func compareGroups(t *testing.T, a, b *rule.Group) {
 	}
 }
 
-func TestManagerUpdateNegative(t *testing.T) {
-	testCases := []struct {
-		notifiers []notifier.Notifier
-		rw        remotewrite.RWClient
-		cfg       config.Group
-		expErr    string
-	}{
-		{
-			nil,
-			nil,
-			config.Group{
-				Name: "Recording rule only",
-				Rules: []config.Rule{
-					{Record: "record", Expr: "max(up)"},
-				},
-			},
-			"contains recording rules",
-		},
-		{
-			nil,
-			nil,
-			config.Group{
-				Name: "Alerting rule only",
-				Rules: []config.Rule{
-					{Alert: "alert", Expr: "up > 0"},
-				},
-			},
-			"contains alerting rules",
-		},
-		{
-			[]notifier.Notifier{&notifier.FakeNotifier{}},
-			nil,
-			config.Group{
-				Name: "Recording and alerting rules",
-				Rules: []config.Rule{
-					{Alert: "alert1", Expr: "up > 0"},
-					{Alert: "alert2", Expr: "up > 0"},
-					{Record: "record", Expr: "max(up)"},
-				},
-			},
-			"contains recording rules",
-		},
-		{
-			nil,
-			&remotewrite.Client{},
-			config.Group{
-				Name: "Recording and alerting rules",
-				Rules: []config.Rule{
-					{Record: "record1", Expr: "max(up)"},
-					{Record: "record2", Expr: "max(up)"},
-					{Alert: "alert", Expr: "up > 0"},
-				},
-			},
-			"contains alerting rules",
-		},
+func TestManagerUpdate_Failure(t *testing.T) {
+	f := func(notifiers []notifier.Notifier, rw remotewrite.RWClient, cfg config.Group, errStrExpected string) {
+		t.Helper()
+
+		m := &manager{
+			groups:         make(map[uint64]*rule.Group),
+			querierBuilder: &datasource.FakeQuerier{},
+			rw:             rw,
+		}
+		if notifiers != nil {
+			m.notifiers = func() []notifier.Notifier { return notifiers }
+		}
+		err := m.update(context.Background(), []config.Group{cfg}, false)
+		if err == nil {
+			t.Fatalf("expected to get error; got nil")
+		}
+		errStr := err.Error()
+		if !strings.Contains(errStr, errStrExpected) {
+			t.Fatalf("missing %q in the error %q", errStrExpected, errStr)
+		}
 	}
 
-	for _, tc := range testCases {
-		t.Run(tc.cfg.Name, func(t *testing.T) {
-			m := &manager{
-				groups:         make(map[uint64]*rule.Group),
-				querierBuilder: &datasource.FakeQuerier{},
-				rw:             tc.rw,
-			}
-			if tc.notifiers != nil {
-				m.notifiers = func() []notifier.Notifier { return tc.notifiers }
-			}
-			err := m.update(context.Background(), []config.Group{tc.cfg}, false)
-			if err == nil {
-				t.Fatalf("expected to get error; got nil")
-			}
-			if !strings.Contains(err.Error(), tc.expErr) {
-				t.Fatalf("expected err to contain %q; got %q", tc.expErr, err)
-			}
-		})
-	}
+	f(nil, nil, config.Group{
+		Name: "Recording rule only",
+		Rules: []config.Rule{
+			{Record: "record", Expr: "max(up)"},
+		},
+	}, "contains recording rules")
+
+	f(nil, nil, config.Group{
+		Name: "Alerting rule only",
+		Rules: []config.Rule{
+			{Alert: "alert", Expr: "up > 0"},
+		},
+	}, "contains alerting rules")
+
+	f([]notifier.Notifier{&notifier.FakeNotifier{}}, nil, config.Group{
+		Name: "Recording and alerting rules",
+		Rules: []config.Rule{
+			{Alert: "alert1", Expr: "up > 0"},
+			{Alert: "alert2", Expr: "up > 0"},
+			{Record: "record", Expr: "max(up)"},
+		},
+	}, "contains recording rules")
+
+	f(nil, &remotewrite.Client{}, config.Group{
+		Name: "Recording and alerting rules",
+		Rules: []config.Rule{
+			{Record: "record1", Expr: "max(up)"},
+			{Record: "record2", Expr: "max(up)"},
+			{Alert: "alert", Expr: "up > 0"},
+		},
+	}, "contains alerting rules")
 }
 
 func loadCfg(t *testing.T, path []string, validateAnnotations, validateExpressions bool) []config.Group {

app/vmalert/multiarch/Dockerfile

@@ -1,7 +1,7 @@
 # See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
 ARG certs_image
 ARG root_image
-FROM $certs_image as certs
+FROM $certs_image AS certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 
 FROM $root_image

app/vmalert/notifier/alert_test.go

@@ -11,7 +11,7 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promrelabel"
 )
 
-func TestAlert_ExecTemplate(t *testing.T) {
+func TestAlertExecTemplate(t *testing.T) {
 	extLabels := make(map[string]string)
 	const (
 		extCluster = "prod"
@@ -23,201 +23,164 @@ func TestAlert_ExecTemplate(t *testing.T) {
 	_, err := Init(nil, extLabels, extURL)
 	checkErr(t, err)
 
-	testCases := []struct {
-		name        string
-		alert       *Alert
-		annotations map[string]string
-		expTpl      map[string]string
-	}{
-		{
-			name:        "empty-alert",
-			alert:       &Alert{},
-			annotations: map[string]string{},
-			expTpl:      map[string]string{},
-		},
-		{
-			name: "no-template",
-			alert: &Alert{
-				Value: 1e4,
-				Labels: map[string]string{
-					"instance": "localhost",
+	f := func(alert *Alert, annotations map[string]string, tplExpected map[string]string) {
+		t.Helper()
+
+		if err := ValidateTemplates(annotations); err != nil {
+			t.Fatalf("cannot validate annotations: %s", err)
+		}
+
+		qFn := func(_ string) ([]datasource.Metric, error) {
+			return []datasource.Metric{
+				{
+					Labels: []datasource.Label{
+						{Name: "foo", Value: "bar"},
+						{Name: "baz", Value: "qux"},
+					},
+					Values:     []float64{1},
+					Timestamps: []int64{1},
 				},
-			},
-			annotations: map[string]string{},
-			expTpl:      map[string]string{},
-		},
-		{
-			name: "label-template",
-			alert: &Alert{
-				Value: 1e4,
-				Labels: map[string]string{
-					"job":      "staging",
-					"instance": "localhost",
+				{
+					Labels: []datasource.Label{
+						{Name: "foo", Value: "garply"},
+						{Name: "baz", Value: "fred"},
+					},
+					Values:     []float64{2},
+					Timestamps: []int64{1},
 				},
-				For: 5 * time.Minute,
-			},
-			annotations: map[string]string{
-				"summary":     "Too high connection number for {{$labels.instance}} for job {{$labels.job}}",
-				"description": "It is {{ $value }} connections for {{$labels.instance}} for more than {{ .For }}",
-			},
-			expTpl: map[string]string{
-				"summary":     "Too high connection number for localhost for job staging",
-				"description": "It is 10000 connections for localhost for more than 5m0s",
-			},
-		},
-		{
-			name: "expression-template",
-			alert: &Alert{
-				Expr: `vm_rows{"label"="bar"}<0`,
-			},
-			annotations: map[string]string{
-				"exprEscapedQuery":  "{{ $expr|queryEscape }}",
-				"exprEscapedPath":   "{{ $expr|pathEscape }}",
-				"exprEscapedJSON":   "{{ $expr|jsonEscape }}",
-				"exprEscapedQuotes": "{{ $expr|quotesEscape }}",
-				"exprEscapedHTML":   "{{ $expr|htmlEscape }}",
-			},
-			expTpl: map[string]string{
-				"exprEscapedQuery":  "vm_rows%7B%22label%22%3D%22bar%22%7D%3C0",
-				"exprEscapedPath":   "vm_rows%7B%22label%22=%22bar%22%7D%3C0",
-				"exprEscapedJSON":   `"vm_rows{\"label\"=\"bar\"}\u003c0"`,
-				"exprEscapedQuotes": `vm_rows{\"label\"=\"bar\"}\u003c0`,
-				"exprEscapedHTML":   "vm_rows{&quot;label&quot;=&quot;bar&quot;}&lt;0",
-			},
-		},
-		{
-			name:  "query",
-			alert: &Alert{Expr: `vm_rows{"label"="bar"}>0`},
-			annotations: map[string]string{
-				"summary": `{{ query "foo" | first | value }}`,
-				"desc":    `{{ range query "bar" }}{{ . | label "foo" }} {{ . | value }};{{ end }}`,
-			},
-			expTpl: map[string]string{
-				"summary": "1",
-				"desc":    "bar 1;garply 2;",
-			},
-		},
-		{
-			name: "external",
-			alert: &Alert{
-				Value: 1e4,
-				Labels: map[string]string{
-					"job":      "staging",
-					"instance": "localhost",
-				},
-			},
-			annotations: map[string]string{
-				"url":         "{{ $externalURL }}",
-				"summary":     "Issues with {{$labels.instance}} (dc-{{$externalLabels.dc}}) for job {{$labels.job}}",
-				"description": "It is {{ $value }} connections for {{$labels.instance}} (cluster-{{$externalLabels.cluster}})",
-			},
-			expTpl: map[string]string{
-				"url":         extURL,
-				"summary":     fmt.Sprintf("Issues with localhost (dc-%s) for job staging", extDC),
-				"description": fmt.Sprintf("It is 10000 connections for localhost (cluster-%s)", extCluster),
-			},
-		},
-		{
-			name: "alert and group IDs",
-			alert: &Alert{
-				ID:      42,
-				GroupID: 24,
-			},
-			annotations: map[string]string{
-				"url": "/api/v1/alert?alertID={{$alertID}}&groupID={{$groupID}}",
-			},
-			expTpl: map[string]string{
-				"url": "/api/v1/alert?alertID=42&groupID=24",
-			},
-		},
-		{
-			name: "ActiveAt time",
-			alert: &Alert{
-				ActiveAt: time.Date(2022, 8, 19, 20, 34, 58, 651387237, time.UTC),
-			},
-			annotations: map[string]string{
-				"diagram": "![](http://example.com?render={{$activeAt.Unix}}",
-			},
-			expTpl: map[string]string{
-				"diagram": "![](http://example.com?render=1660941298",
-			},
-		},
-		{
-			name:  "ActiveAt time is nil",
-			alert: &Alert{},
-			annotations: map[string]string{
-				"default_time": "{{$activeAt}}",
-			},
-			expTpl: map[string]string{
-				"default_time": "0001-01-01 00:00:00 +0000 UTC",
-			},
-		},
-		{
-			name: "ActiveAt custom format",
-			alert: &Alert{
-				ActiveAt: time.Date(2022, 8, 19, 20, 34, 58, 651387237, time.UTC),
-			},
-			annotations: map[string]string{
-				"fire_time": `{{$activeAt.Format "2006/01/02 15:04:05"}}`,
-			},
-			expTpl: map[string]string{
-				"fire_time": "2022/08/19 20:34:58",
-			},
-		},
-		{
-			name: "ActiveAt query range",
-			alert: &Alert{
-				ActiveAt: time.Date(2022, 8, 19, 20, 34, 58, 651387237, time.UTC),
-			},
-			annotations: map[string]string{
-				"grafana_url": `vm-grafana.com?from={{($activeAt.Add (parseDurationTime "1h")).Unix}}&to={{($activeAt.Add (parseDurationTime "-1h")).Unix}}`,
-			},
-			expTpl: map[string]string{
-				"grafana_url": "vm-grafana.com?from=1660944898&to=1660937698",
-			},
-		},
+			}, nil
+		}
+
+		tpl, err := alert.ExecTemplate(qFn, alert.Labels, annotations)
+		if err != nil {
+			t.Fatalf("cannot execute template: %s", err)
+		}
+		if len(tpl) != len(tplExpected) {
+			t.Fatalf("unexpected number of elements; got %d; want %d", len(tpl), len(tplExpected))
+		}
+		for k := range tplExpected {
+			got, exp := tpl[k], tplExpected[k]
+			if got != exp {
+				t.Fatalf("unexpected template for key=%q; got %q; want %q", k, got, exp)
+			}
+		}
 	}
 
-	qFn := func(_ string) ([]datasource.Metric, error) {
-		return []datasource.Metric{
-			{
-				Labels: []datasource.Label{
-					{Name: "foo", Value: "bar"},
-					{Name: "baz", Value: "qux"},
-				},
-				Values:     []float64{1},
-				Timestamps: []int64{1},
-			},
-			{
-				Labels: []datasource.Label{
-					{Name: "foo", Value: "garply"},
-					{Name: "baz", Value: "fred"},
-				},
-				Values:     []float64{2},
-				Timestamps: []int64{1},
-			},
-		}, nil
-	}
-	for _, tc := range testCases {
-		t.Run(tc.name, func(t *testing.T) {
-			if err := ValidateTemplates(tc.annotations); err != nil {
-				t.Fatal(err)
-			}
-			tpl, err := tc.alert.ExecTemplate(qFn, tc.alert.Labels, tc.annotations)
-			if err != nil {
-				t.Fatal(err)
-			}
-			if len(tpl) != len(tc.expTpl) {
-				t.Fatalf("expected %d elements; got %d", len(tc.expTpl), len(tpl))
-			}
-			for k := range tc.expTpl {
-				got, exp := tpl[k], tc.expTpl[k]
-				if got != exp {
-					t.Fatalf("expected %q=%q; got %q=%q", k, exp, k, got)
-				}
-			}
-		})
-	}
+	// empty-alert
+	f(&Alert{}, map[string]string{}, map[string]string{})
+
+	// no-template
+	f(&Alert{
+		Value: 1e4,
+		Labels: map[string]string{
+			"instance": "localhost",
+		},
+	}, map[string]string{}, map[string]string{})
+
+	// label-template
+	f(&Alert{
+		Value: 1e4,
+		Labels: map[string]string{
+			"job":      "staging",
+			"instance": "localhost",
+		},
+		For: 5 * time.Minute,
+	}, map[string]string{
+		"summary":     "Too high connection number for {{$labels.instance}} for job {{$labels.job}}",
+		"description": "It is {{ $value }} connections for {{$labels.instance}} for more than {{ .For }}",
+	}, map[string]string{
+		"summary":     "Too high connection number for localhost for job staging",
+		"description": "It is 10000 connections for localhost for more than 5m0s",
+	})
+
+	// expression-template
+	f(&Alert{
+		Expr: `vm_rows{"label"="bar"}<0`,
+	}, map[string]string{
+		"exprEscapedQuery":  "{{ $expr|queryEscape }}",
+		"exprEscapedPath":   "{{ $expr|pathEscape }}",
+		"exprEscapedJSON":   "{{ $expr|jsonEscape }}",
+		"exprEscapedQuotes": "{{ $expr|quotesEscape }}",
+		"exprEscapedHTML":   "{{ $expr|htmlEscape }}",
+	}, map[string]string{
+		"exprEscapedQuery":  "vm_rows%7B%22label%22%3D%22bar%22%7D%3C0",
+		"exprEscapedPath":   "vm_rows%7B%22label%22=%22bar%22%7D%3C0",
+		"exprEscapedJSON":   `"vm_rows{\"label\"=\"bar\"}\u003c0"`,
+		"exprEscapedQuotes": `vm_rows{\"label\"=\"bar\"}\u003c0`,
+		"exprEscapedHTML":   "vm_rows{&quot;label&quot;=&quot;bar&quot;}&lt;0",
+	})
+
+	// query
+	f(&Alert{
+		Expr: `vm_rows{"label"="bar"}>0`,
+	}, map[string]string{
+		"summary": `{{ query "foo" | first | value }}`,
+		"desc":    `{{ range query "bar" }}{{ . | label "foo" }} {{ . | value }};{{ end }}`,
+	}, map[string]string{
+		"summary": "1",
+		"desc":    "bar 1;garply 2;",
+	})
+
+	// external
+	f(&Alert{
+		Value: 1e4,
+		Labels: map[string]string{
+			"job":      "staging",
+			"instance": "localhost",
+		},
+	}, map[string]string{
+		"url":         "{{ $externalURL }}",
+		"summary":     "Issues with {{$labels.instance}} (dc-{{$externalLabels.dc}}) for job {{$labels.job}}",
+		"description": "It is {{ $value }} connections for {{$labels.instance}} (cluster-{{$externalLabels.cluster}})",
+	}, map[string]string{
+		"url":         extURL,
+		"summary":     fmt.Sprintf("Issues with localhost (dc-%s) for job staging", extDC),
+		"description": fmt.Sprintf("It is 10000 connections for localhost (cluster-%s)", extCluster),
+	})
+
+	// alert and group IDs
+	f(&Alert{
+		ID:      42,
+		GroupID: 24,
+	}, map[string]string{
+		"url": "/api/v1/alert?alertID={{$alertID}}&groupID={{$groupID}}",
+	}, map[string]string{
+		"url": "/api/v1/alert?alertID=42&groupID=24",
+	})
+
+	// ActiveAt time
+	f(&Alert{
+		ActiveAt: time.Date(2022, 8, 19, 20, 34, 58, 651387237, time.UTC),
+	}, map[string]string{
+		"diagram": "![](http://example.com?render={{$activeAt.Unix}}",
+	}, map[string]string{
+		"diagram": "![](http://example.com?render=1660941298",
+	})
+
+	// ActiveAt time is nil
+	f(&Alert{}, map[string]string{
+		"default_time": "{{$activeAt}}",
+	}, map[string]string{
+		"default_time": "0001-01-01 00:00:00 +0000 UTC",
+	})
+
+	// ActiveAt custom format
+	f(&Alert{
+		ActiveAt: time.Date(2022, 8, 19, 20, 34, 58, 651387237, time.UTC),
+	}, map[string]string{
+		"fire_time": `{{$activeAt.Format "2006/01/02 15:04:05"}}`,
+	}, map[string]string{
+		"fire_time": "2022/08/19 20:34:58",
+	})
+
+	// ActiveAt query range
+	f(&Alert{
+		ActiveAt: time.Date(2022, 8, 19, 20, 34, 58, 651387237, time.UTC),
+	}, map[string]string{
+		"grafana_url": `vm-grafana.com?from={{($activeAt.Add (parseDurationTime "1h")).Unix}}&to={{($activeAt.Add (parseDurationTime "-1h")).Unix}}`,
+	}, map[string]string{
+		"grafana_url": "vm-grafana.com?from=1660944898&to=1660937698",
+	})
 }
 
 func TestAlert_toPromLabels(t *testing.T) {

app/vmalert/notifier/alertmanager_test.go

@@ -16,10 +16,10 @@ func TestAlertManager_Addr(t *testing.T) {
 	const addr = "http://localhost"
 	am, err := NewAlertManager(addr, nil, promauth.HTTPClientConfig{}, nil, 0)
 	if err != nil {
-		t.Errorf("unexpected error: %s", err)
+		t.Fatalf("unexpected error: %s", err)
 	}
 	if am.Addr() != addr {
-		t.Errorf("expected to have %q; got %q", addr, am.Addr())
+		t.Fatalf("expected to have %q; got %q", addr, am.Addr())
 	}
 }
 
@@ -28,21 +28,20 @@ func TestAlertManager_Send(t *testing.T) {
 	const headerKey, headerValue = "TenantID", "foo"
 	mux := http.NewServeMux()
 	mux.HandleFunc("/", func(_ http.ResponseWriter, _ *http.Request) {
-		t.Errorf("should not be called")
+		t.Fatalf("should not be called")
 	})
 	c := -1
 	mux.HandleFunc(alertManagerPath, func(w http.ResponseWriter, r *http.Request) {
 		user, pass, ok := r.BasicAuth()
 		if !ok {
-			t.Errorf("unauthorized request")
+			t.Fatalf("unauthorized request")
 		}
 		if user != baUser || pass != baPass {
-			t.Errorf("wrong creds %q:%q; expected %q:%q",
-				user, pass, baUser, baPass)
+			t.Fatalf("wrong creds %q:%q; expected %q:%q", user, pass, baUser, baPass)
 		}
 		c++
 		if r.Method != http.MethodPost {
-			t.Errorf("expected POST method got %s", r.Method)
+			t.Fatalf("expected POST method got %s", r.Method)
 		}
 		switch c {
 		case 0:
@@ -59,25 +58,23 @@ func TestAlertManager_Send(t *testing.T) {
 				GeneratorURL string            `json:"generatorURL"`
 			}
 			if err := json.NewDecoder(r.Body).Decode(&a); err != nil {
-				t.Errorf("can not unmarshal data into alert %s", err)
-				t.FailNow()
+				t.Fatalf("can not unmarshal data into alert %s", err)
 			}
 			if len(a) != 1 {
-				t.Errorf("expected 1 alert in array got %d", len(a))
+				t.Fatalf("expected 1 alert in array got %d", len(a))
 			}
 			if a[0].GeneratorURL != "0/0" {
-				t.Errorf("expected 0/0 as generatorURL got %s", a[0].GeneratorURL)
+				t.Fatalf("expected 0/0 as generatorURL got %s", a[0].GeneratorURL)
 			}
 			if a[0].StartsAt.IsZero() {
-				t.Errorf("expected non-zero start time")
+				t.Fatalf("expected non-zero start time")
 			}
 			if a[0].EndAt.IsZero() {
-				t.Errorf("expected non-zero end time")
+				t.Fatalf("expected non-zero end time")
 			}
 		case 3:
 			if r.Header.Get(headerKey) != headerValue {
-				t.Errorf("expected header %q to be set to %q; got %q instead",
-					headerKey, headerValue, r.Header.Get(headerKey))
+				t.Fatalf("expected header %q to be set to %q; got %q instead", headerKey, headerValue, r.Header.Get(headerKey))
 			}
 		}
 	})
@@ -94,13 +91,13 @@ func TestAlertManager_Send(t *testing.T) {
 		return strconv.FormatUint(alert.GroupID, 10) + "/" + strconv.FormatUint(alert.ID, 10)
 	}, aCfg, nil, 0)
 	if err != nil {
-		t.Errorf("unexpected error: %s", err)
+		t.Fatalf("unexpected error: %s", err)
 	}
 	if err := am.Send(context.Background(), []Alert{{}, {}}, nil); err == nil {
-		t.Error("expected connection error got nil")
+		t.Fatalf("expected connection error got nil")
 	}
 	if err := am.Send(context.Background(), []Alert{}, nil); err == nil {
-		t.Error("expected wrong http code error got nil")
+		t.Fatalf("expected wrong http code error got nil")
 	}
 	if err := am.Send(context.Background(), []Alert{{
 		GroupID:     0,
@@ -109,12 +106,12 @@ func TestAlertManager_Send(t *testing.T) {
 		End:         time.Now().UTC(),
 		Annotations: map[string]string{"a": "b", "c": "d", "e": "f"},
 	}}, nil); err != nil {
-		t.Errorf("unexpected error %s", err)
+		t.Fatalf("unexpected error %s", err)
 	}
 	if c != 2 {
-		t.Errorf("expected 2 calls(count from zero) to server got %d", c)
+		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 {
-		t.Errorf("unexpected error %s", err)
+		t.Fatalf("unexpected error %s", err)
 	}
 }

app/vmalert/notifier/config.go

@@ -51,7 +51,7 @@ type Config struct {
 	Checksum string
 
 	// Catches all undefined fields and must be empty after parsing.
-	XXX map[string]interface{} `yaml:",inline"`
+	XXX map[string]any `yaml:",inline"`
 
 	// This is set to the directory from where the config has been loaded.
 	baseDir string
@@ -73,7 +73,7 @@ type StaticConfig struct {
 }
 
 // UnmarshalYAML implements the yaml.Unmarshaler interface.
-func (cfg *Config) UnmarshalYAML(unmarshal func(interface{}) error) error {
+func (cfg *Config) UnmarshalYAML(unmarshal func(any) error) error {
 	type config Config
 	if err := unmarshal((*config)(cfg)); err != nil {
 		return err

app/vmalert/notifier/config_test.go

@@ -5,10 +5,14 @@ import (
 	"testing"
 )
 
-func TestConfigParseGood(t *testing.T) {
+func TestParseConfig_Success(t *testing.T) {
 	f := func(path string) {
+		t.Helper()
+
 		_, err := parseConfig(path)
-		checkErr(t, err)
+		if err != nil {
+			t.Fatalf("unexpected error: %s", err)
+		}
 	}
 	f("testdata/mixed.good.yaml")
 	f("testdata/consul.good.yaml")
@@ -16,14 +20,16 @@ func TestConfigParseGood(t *testing.T) {
 	f("testdata/static.good.yaml")
 }
 
-func TestConfigParseBad(t *testing.T) {
+func TestParseConfig_Failure(t *testing.T) {
 	f := func(path, expErr string) {
+		t.Helper()
+
 		_, err := parseConfig(path)
 		if err == nil {
 			t.Fatalf("expected to get non-nil err for config %q", path)
 		}
 		if !strings.Contains(err.Error(), expErr) {
-			t.Errorf("expected err to contain %q; got %q instead", expErr, err)
+			t.Fatalf("expected err to contain %q; got %q instead", expErr, err)
 		}
 	}
 

app/vmalert/notifier/config_watcher_test.go

@@ -319,46 +319,41 @@ func TestMergeHTTPClientConfigs(t *testing.T) {
 	}
 }
 
-func TestParseLabels(t *testing.T) {
-	testCases := []struct {
-		name            string
-		target          string
-		cfg             *Config
-		expectedAddress string
-		expectedErr     bool
-	}{
-		{
-			"invalid address",
-			"invalid:*//url",
-			&Config{},
-			"",
-			true,
-		},
-		{
-			"use some default params",
-			"alertmanager:9093",
-			&Config{PathPrefix: "test"},
-			"http://alertmanager:9093/test/api/v2/alerts",
-			false,
-		},
-		{
-			"use target address",
-			"https://alertmanager:9093/api/v1/alerts",
-			&Config{Scheme: "http", PathPrefix: "test"},
-			"https://alertmanager:9093/api/v1/alerts",
-			false,
-		},
+func TestParseLabels_Failure(t *testing.T) {
+	f := func(target string, cfg *Config) {
+		t.Helper()
+
+		_, _, err := parseLabels(target, nil, cfg)
+		if err == nil {
+			t.Fatalf("expecting non-nil error")
+		}
 	}
 
-	for _, tc := range testCases {
-		t.Run(tc.name, func(t *testing.T) {
-			address, _, err := parseLabels(tc.target, nil, tc.cfg)
-			if err == nil == tc.expectedErr {
-				t.Fatalf("unexpected error; got %t; want %t", err != nil, tc.expectedErr)
-			}
-			if address != tc.expectedAddress {
-				t.Fatalf("unexpected address; got %q; want %q", address, tc.expectedAddress)
-			}
-		})
-	}
+	// invalid address
+	f("invalid:*//url", &Config{})
+}
+
+func TestParseLabels_Success(t *testing.T) {
+	f := func(target string, cfg *Config, expectedAddress string) {
+		t.Helper()
+
+		address, _, err := parseLabels(target, nil, cfg)
+		if err != nil {
+			t.Fatalf("unexpected error: %s", err)
+		}
+		if address != expectedAddress {
+			t.Fatalf("unexpected address; got %q; want %q", address, expectedAddress)
+		}
+	}
+
+	// use some default params
+	f("alertmanager:9093", &Config{
+		PathPrefix: "test",
+	}, "http://alertmanager:9093/test/api/v2/alerts")
+
+	// use target address
+	f("https://alertmanager:9093/api/v1/alerts", &Config{
+		Scheme:     "http",
+		PathPrefix: "test",
+	}, "https://alertmanager:9093/api/v1/alerts")
 }

app/vmalert/notifier/notifier_blackhole_test.go

@@ -17,12 +17,12 @@ func TestBlackHoleNotifier_Send(t *testing.T) {
 		End:         time.Now().UTC(),
 		Annotations: map[string]string{"a": "b", "c": "d", "e": "f"},
 	}}, nil); err != nil {
-		t.Errorf("unexpected error %s", err)
+		t.Fatalf("unexpected error %s", err)
 	}
 
 	alertCount := bh.metrics.alertsSent.Get()
 	if alertCount != 1 {
-		t.Errorf("expect value 1; instead got %d", alertCount)
+		t.Fatalf("expect value 1; instead got %d", alertCount)
 	}
 }
 
@@ -35,7 +35,7 @@ func TestBlackHoleNotifier_Close(t *testing.T) {
 		End:         time.Now().UTC(),
 		Annotations: map[string]string{"a": "b", "c": "d", "e": "f"},
 	}}, nil); err != nil {
-		t.Errorf("unexpected error %s", err)
+		t.Fatalf("unexpected error %s", err)
 	}
 
 	bh.Close()
@@ -44,7 +44,7 @@ func TestBlackHoleNotifier_Close(t *testing.T) {
 	alertMetricName := "vmalert_alerts_sent_total{addr=\"blackhole\"}"
 	for _, name := range defaultMetrics.ListMetricNames() {
 		if name == alertMetricName {
-			t.Errorf("Metric name should have unregistered.But still present")
+			t.Fatalf("Metric name should have unregistered.But still present")
 		}
 	}
 }

app/vmalert/remoteread/init.go

@@ -10,6 +10,7 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/utils"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputils"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/netutil"
 )
 
 var (
@@ -70,7 +71,7 @@ func Init() (datasource.QuerierBuilder, error) {
 		return nil, fmt.Errorf("failed to create transport: %w", err)
 	}
 	tr.IdleConnTimeout = *idleConnectionTimeout
-	tr.DialContext = httputils.GetStatDialFunc("vmalert_remoteread")
+	tr.DialContext = netutil.NewStatDialFunc("vmalert_remoteread")
 
 	endpointParams, err := flagutil.ParseJSONMap(*oauth2EndpointParams)
 	if err != nil {

app/vmalert/remotewrite/client.go

@@ -212,6 +212,9 @@ var (
 	})
 )
 
+// GetDroppedRows returns value of droppedRows metric
+func GetDroppedRows() int64 { return int64(droppedRows.Get()) }
+
 // flush is a blocking function that marshals WriteRequest and sends
 // it to remote-write endpoint. Flush performs limited amount of retries
 // if request fails.

app/vmalert/remotewrite/client_test.go

@@ -44,8 +44,7 @@ func TestClient_Push(t *testing.T) {
 	}
 
 	r := rand.New(rand.NewSource(1))
-	const rowsN = 1e4
-	var sent int
+	const rowsN = int(1e4)
 	for i := 0; i < rowsN; i++ {
 		s := prompbmarshal.TimeSeries{
 			Samples: []prompbmarshal.Sample{{
@@ -57,17 +56,11 @@ func TestClient_Push(t *testing.T) {
 		if err != nil {
 			t.Fatalf("unexpected err: %s", err)
 		}
-		if err == nil {
-			sent++
-		}
 		err = faultyClient.Push(s)
 		if err != nil {
 			t.Fatalf("unexpected err: %s", err)
 		}
 	}
-	if sent == 0 {
-		t.Fatalf("0 series sent")
-	}
 	if err := client.Close(); err != nil {
 		t.Fatalf("failed to close client: %s", err)
 	}
@@ -75,77 +68,66 @@ func TestClient_Push(t *testing.T) {
 		t.Fatalf("failed to close faulty client: %s", err)
 	}
 	got := testSrv.accepted()
-	if got != sent {
-		t.Fatalf("expected to have %d series; got %d", sent, got)
+	if got != rowsN {
+		t.Fatalf("expected to have %d series; got %d", rowsN, got)
 	}
 	got = faultySrv.accepted()
-	if got != sent {
-		t.Fatalf("expected to have %d series for faulty client; got %d", sent, got)
+	if got != rowsN {
+		t.Fatalf("expected to have %d series for faulty client; got %d", rowsN, got)
 	}
 }
 
 func TestClient_run_maxBatchSizeDuringShutdown(t *testing.T) {
-	batchSize := 20
+	const batchSize = 20
 
-	testTable := []struct {
-		name     string // name of the test case
-		pushCnt  int    // how many time series is pushed to the client
-		batchCnt int    // the expected batch count sent by the client
-	}{
-		{
-			name:     "pushCnt % batchSize == 0",
-			pushCnt:  batchSize * 40,
-			batchCnt: 40,
-		},
-		{
-			name:     "pushCnt % batchSize != 0",
-			pushCnt:  batchSize*40 + 1,
-			batchCnt: 40 + 1,
-		},
-	}
+	f := func(pushCnt, batchCntExpected int) {
+		t.Helper()
 
-	for _, tt := range testTable {
-		t.Run(tt.name, func(t *testing.T) {
-			// run new server
-			bcServer := newBatchCntRWServer()
+		// run new server
+		bcServer := newBatchCntRWServer()
 
-			// run new client
-			rwClient, err := NewClient(context.Background(), Config{
-				MaxBatchSize: batchSize,
+		// run new client
+		rwClient, err := NewClient(context.Background(), Config{
+			MaxBatchSize: batchSize,
 
-				// Set everything to 1 to simplify the calculation.
-				Concurrency:   1,
-				MaxQueueSize:  1000,
-				FlushInterval: time.Minute,
+			// Set everything to 1 to simplify the calculation.
+			Concurrency:   1,
+			MaxQueueSize:  1000,
+			FlushInterval: time.Minute,
 
-				// batch count server
-				Addr: bcServer.URL,
-			})
-			if err != nil {
-				t.Fatalf("new remote write client failed, err: %v", err)
-			}
-
-			// push time series to the client.
-			for i := 0; i < tt.pushCnt; i++ {
-				if err = rwClient.Push(prompbmarshal.TimeSeries{}); err != nil {
-					t.Fatalf("push time series to the client failed, err: %v", err)
-				}
-			}
-
-			// close the client so the rest ts will be flushed in `shutdown`
-			if err = rwClient.Close(); err != nil {
-				t.Fatalf("shutdown client failed, err: %v", err)
-			}
-
-			// finally check how many batches is sent.
-			if tt.batchCnt != bcServer.acceptedBatches() {
-				t.Errorf("client sent batch count incorrect, want: %d, get: %d", tt.batchCnt, bcServer.acceptedBatches())
-			}
-			if tt.pushCnt != bcServer.accepted() {
-				t.Errorf("client sent time series count incorrect, want: %d, get: %d", tt.pushCnt, bcServer.accepted())
-			}
+			// batch count server
+			Addr: bcServer.URL,
 		})
+		if err != nil {
+			t.Fatalf("cannot create remote write client: %s", err)
+		}
+
+		// push time series to the client.
+		for i := 0; i < pushCnt; i++ {
+			if err = rwClient.Push(prompbmarshal.TimeSeries{}); err != nil {
+				t.Fatalf("cannot time series to the client: %s", err)
+			}
+		}
+
+		// close the client so the rest ts will be flushed in `shutdown`
+		if err = rwClient.Close(); err != nil {
+			t.Fatalf("cannot shutdown client: %s", err)
+		}
+
+		// finally check how many batches is sent.
+		if bcServer.acceptedBatches() != batchCntExpected {
+			t.Fatalf("client sent batch count incorrect; got %d; want %d", bcServer.acceptedBatches(), batchCntExpected)
+		}
+		if pushCnt != bcServer.accepted() {
+			t.Fatalf("client sent time series count incorrect; got %d; want %d", bcServer.accepted(), pushCnt)
+		}
 	}
+
+	// pushCnt % batchSize == 0
+	f(batchSize*40, 40)
+
+	//pushCnt % batchSize != 0
+	f(batchSize*40+1, 40+1)
 }
 
 func newRWServer() *rwServer {

app/vmalert/remotewrite/init.go

@@ -9,6 +9,7 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/utils"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputils"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/netutil"
 )
 
 var (
@@ -74,7 +75,7 @@ func Init(ctx context.Context) (*Client, error) {
 		return nil, fmt.Errorf("failed to create transport: %w", err)
 	}
 	t.IdleConnTimeout = *idleConnectionTimeout
-	t.DialContext = httputils.GetStatDialFunc("vmalert_remotewrite")
+	t.DialContext = netutil.NewStatDialFunc("vmalert_remotewrite")
 
 	endpointParams, err := flagutil.ParseJSONMap(*oauth2EndpointParams)
 	if err != nil {

app/vmalert/replay.go

@@ -15,9 +15,10 @@ import (
 
 var (
 	replayFrom = flag.String("replay.timeFrom", "",
-		"The time filter in RFC3339 format to select time series with timestamp equal or higher than provided value. E.g. '2020-01-01T20:07:00Z'")
+		"The time filter in RFC3339 format to start the replay from. E.g. '2020-01-01T20:07:00Z'")
 	replayTo = flag.String("replay.timeTo", "",
-		"The time filter in RFC3339 format to select timeseries with timestamp equal or lower than provided value. E.g. '2020-01-01T20:07:00Z'")
+		"The time filter in RFC3339 format to finish the replay by. E.g. '2020-01-01T20:07:00Z'. "+
+			"By default, is set to the current time.")
 	replayRulesDelay = flag.Duration("replay.rulesDelay", time.Second,
 		"Delay between rules evaluation within the group. Could be important if there are chained rules inside the group "+
 			"and processing need to wait for previous rule results to be persisted by remote storage before evaluating the next rule."+
@@ -36,14 +37,20 @@ func replay(groupsCfg []config.Group, qb datasource.QuerierBuilder, rw remotewri
 	}
 	tFrom, err := time.Parse(time.RFC3339, *replayFrom)
 	if err != nil {
-		return fmt.Errorf("failed to parse %q: %w", *replayFrom, err)
+		return fmt.Errorf("failed to parse replay.timeFrom=%q: %w", *replayFrom, err)
 	}
-	tTo, err := time.Parse(time.RFC3339, *replayTo)
-	if err != nil {
-		return fmt.Errorf("failed to parse %q: %w", *replayTo, err)
+
+	// use tFrom location for default value, otherwise filters could have different locations
+	tTo := time.Now().In(tFrom.Location())
+	if *replayTo != "" {
+		tTo, err = time.Parse(time.RFC3339, *replayTo)
+		if err != nil {
+			return fmt.Errorf("failed to parse replay.timeTo=%q: %w", *replayTo, err)
+		}
 	}
+
 	if !tTo.After(tFrom) {
-		return fmt.Errorf("replay.timeTo must be bigger than replay.timeFrom")
+		return fmt.Errorf("replay.timeTo=%v must be bigger than replay.timeFrom=%v", tTo, tFrom)
 	}
 	labels := make(map[string]string)
 	for _, s := range *externalLabels {
@@ -68,9 +75,13 @@ func replay(groupsCfg []config.Group, qb datasource.QuerierBuilder, rw remotewri
 		ng := rule.NewGroup(cfg, qb, *evaluationInterval, labels)
 		total += ng.Replay(tFrom, tTo, rw, *replayMaxDatapoints, *replayRuleRetryAttempts, *replayRulesDelay, *disableProgressBar)
 	}
-	logger.Infof("replay finished! Imported %d samples", total)
-	if rw != nil {
-		return rw.Close()
+	logger.Infof("replay evaluation finished, generated %d samples", total)
+	if err := rw.Close(); err != nil {
+		return err
+	}
+	droppedRows := remotewrite.GetDroppedRows()
+	if droppedRows > 0 {
+		return fmt.Errorf("failed to push all generated samples to remote write url, dropped %d samples out of %d", droppedRows, total)
 	}
 	return nil
 }

app/vmalert/replay_test.go

@@ -8,6 +8,7 @@ import (
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/config"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/datasource"
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/remotewrite"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promutils"
 )
 
@@ -38,135 +39,102 @@ func (fr *fakeReplayQuerier) QueryRange(_ context.Context, q string, from, to ti
 }
 
 func TestReplay(t *testing.T) {
-	testCases := []struct {
-		name     string
-		from, to string
-		maxDP    int
-		cfg      []config.Group
-		qb       *fakeReplayQuerier
-	}{
-		{
-			name:  "one rule + one response",
-			from:  "2021-01-01T12:00:00.000Z",
-			to:    "2021-01-01T12:02:00.000Z",
-			maxDP: 10,
-			cfg: []config.Group{
-				{Rules: []config.Rule{{Record: "foo", Expr: "sum(up)"}}},
-			},
-			qb: &fakeReplayQuerier{
-				registry: map[string]map[string]struct{}{
-					"sum(up)": {"12:00:00+12:02:00": {}},
-				},
-			},
-		},
-		{
-			name:  "one rule + multiple responses",
-			from:  "2021-01-01T12:00:00.000Z",
-			to:    "2021-01-01T12:02:30.000Z",
-			maxDP: 1,
-			cfg: []config.Group{
-				{Rules: []config.Rule{{Record: "foo", Expr: "sum(up)"}}},
-			},
-			qb: &fakeReplayQuerier{
-				registry: map[string]map[string]struct{}{
-					"sum(up)": {
-						"12:00:00+12:01:00": {},
-						"12:01:00+12:02:00": {},
-						"12:02:00+12:02:30": {},
-					},
-				},
-			},
-		},
-		{
-			name:  "datapoints per step",
-			from:  "2021-01-01T12:00:00.000Z",
-			to:    "2021-01-01T15:02:30.000Z",
-			maxDP: 60,
-			cfg: []config.Group{
-				{Interval: promutils.NewDuration(time.Minute), Rules: []config.Rule{{Record: "foo", Expr: "sum(up)"}}},
-			},
-			qb: &fakeReplayQuerier{
-				registry: map[string]map[string]struct{}{
-					"sum(up)": {
-						"12:00:00+13:00:00": {},
-						"13:00:00+14:00:00": {},
-						"14:00:00+15:00:00": {},
-						"15:00:00+15:02:30": {},
-					},
-				},
-			},
-		},
-		{
-			name:  "multiple recording rules + multiple responses",
-			from:  "2021-01-01T12:00:00.000Z",
-			to:    "2021-01-01T12:02:30.000Z",
-			maxDP: 1,
-			cfg: []config.Group{
-				{Rules: []config.Rule{{Record: "foo", Expr: "sum(up)"}}},
-				{Rules: []config.Rule{{Record: "bar", Expr: "max(up)"}}},
-			},
-			qb: &fakeReplayQuerier{
-				registry: map[string]map[string]struct{}{
-					"sum(up)": {
-						"12:00:00+12:01:00": {},
-						"12:01:00+12:02:00": {},
-						"12:02:00+12:02:30": {},
-					},
-					"max(up)": {
-						"12:00:00+12:01:00": {},
-						"12:01:00+12:02:00": {},
-						"12:02:00+12:02:30": {},
-					},
-				},
-			},
-		},
-		{
-			name:  "multiple alerting rules + multiple responses",
-			from:  "2021-01-01T12:00:00.000Z",
-			to:    "2021-01-01T12:02:30.000Z",
-			maxDP: 1,
-			cfg: []config.Group{
-				{Rules: []config.Rule{{Alert: "foo", Expr: "sum(up) > 1"}}},
-				{Rules: []config.Rule{{Alert: "bar", Expr: "max(up) < 1"}}},
-			},
-			qb: &fakeReplayQuerier{
-				registry: map[string]map[string]struct{}{
-					"sum(up) > 1": {
-						"12:00:00+12:01:00": {},
-						"12:01:00+12:02:00": {},
-						"12:02:00+12:02:30": {},
-					},
-					"max(up) < 1": {
-						"12:00:00+12:01:00": {},
-						"12:01:00+12:02:00": {},
-						"12:02:00+12:02:30": {},
-					},
-				},
-			},
-		},
+	f := func(from, to string, maxDP int, cfg []config.Group, qb *fakeReplayQuerier) {
+		t.Helper()
+
+		fromOrig, toOrig, maxDatapointsOrig := *replayFrom, *replayTo, *replayMaxDatapoints
+		retriesOrig, delayOrig := *replayRuleRetryAttempts, *replayRulesDelay
+		defer func() {
+			*replayFrom, *replayTo = fromOrig, toOrig
+			*replayMaxDatapoints, *replayRuleRetryAttempts = maxDatapointsOrig, retriesOrig
+			*replayRulesDelay = delayOrig
+		}()
+
+		*replayRuleRetryAttempts = 1
+		*replayRulesDelay = time.Millisecond
+		rwb := &remotewrite.DebugClient{}
+		*replayFrom = from
+		*replayTo = to
+		*replayMaxDatapoints = maxDP
+		if err := replay(cfg, qb, rwb); err != nil {
+			t.Fatalf("replay failed: %s", err)
+		}
+		if len(qb.registry) > 0 {
+			t.Fatalf("not all requests were sent: %#v", qb.registry)
+		}
 	}
 
-	from, to, maxDP := *replayFrom, *replayTo, *replayMaxDatapoints
-	retries, delay := *replayRuleRetryAttempts, *replayRulesDelay
-	defer func() {
-		*replayFrom, *replayTo = from, to
-		*replayMaxDatapoints, *replayRuleRetryAttempts = maxDP, retries
-		*replayRulesDelay = delay
-	}()
+	// one rule + one response
+	f("2021-01-01T12:00:00.000Z", "2021-01-01T12:02:00.000Z", 10, []config.Group{
+		{Rules: []config.Rule{{Record: "foo", Expr: "sum(up)"}}},
+	}, &fakeReplayQuerier{
+		registry: map[string]map[string]struct{}{
+			"sum(up)": {"12:00:00+12:02:00": {}},
+		},
+	})
 
-	*replayRuleRetryAttempts = 1
-	*replayRulesDelay = time.Millisecond
-	for _, tc := range testCases {
-		t.Run(tc.name, func(t *testing.T) {
-			*replayFrom = tc.from
-			*replayTo = tc.to
-			*replayMaxDatapoints = tc.maxDP
-			if err := replay(tc.cfg, tc.qb, nil); err != nil {
-				t.Fatalf("replay failed: %s", err)
-			}
-			if len(tc.qb.registry) > 0 {
-				t.Fatalf("not all requests were sent: %#v", tc.qb.registry)
-			}
-		})
-	}
+	// one rule + multiple responses
+	f("2021-01-01T12:00:00.000Z", "2021-01-01T12:02:30.000Z", 1, []config.Group{
+		{Rules: []config.Rule{{Record: "foo", Expr: "sum(up)"}}},
+	}, &fakeReplayQuerier{
+		registry: map[string]map[string]struct{}{
+			"sum(up)": {
+				"12:00:00+12:01:00": {},
+				"12:01:00+12:02:00": {},
+				"12:02:00+12:02:30": {},
+			},
+		},
+	})
+
+	// datapoints per step
+	f("2021-01-01T12:00:00.000Z", "2021-01-01T15:02:30.000Z", 60, []config.Group{
+		{Interval: promutils.NewDuration(time.Minute), Rules: []config.Rule{{Record: "foo", Expr: "sum(up)"}}},
+	}, &fakeReplayQuerier{
+		registry: map[string]map[string]struct{}{
+			"sum(up)": {
+				"12:00:00+13:00:00": {},
+				"13:00:00+14:00:00": {},
+				"14:00:00+15:00:00": {},
+				"15:00:00+15:02:30": {},
+			},
+		},
+	})
+
+	// multiple recording rules + multiple responses
+	f("2021-01-01T12:00:00.000Z", "2021-01-01T12:02:30.000Z", 1, []config.Group{
+		{Rules: []config.Rule{{Record: "foo", Expr: "sum(up)"}}},
+		{Rules: []config.Rule{{Record: "bar", Expr: "max(up)"}}},
+	}, &fakeReplayQuerier{
+		registry: map[string]map[string]struct{}{
+			"sum(up)": {
+				"12:00:00+12:01:00": {},
+				"12:01:00+12:02:00": {},
+				"12:02:00+12:02:30": {},
+			},
+			"max(up)": {
+				"12:00:00+12:01:00": {},
+				"12:01:00+12:02:00": {},
+				"12:02:00+12:02:30": {},
+			},
+		},
+	})
+
+	// multiple alerting rules + multiple responses
+	f("2021-01-01T12:00:00.000Z", "2021-01-01T12:02:30.000Z", 1, []config.Group{
+		{Rules: []config.Rule{{Alert: "foo", Expr: "sum(up) > 1"}}},
+		{Rules: []config.Rule{{Alert: "bar", Expr: "max(up) < 1"}}},
+	}, &fakeReplayQuerier{
+		registry: map[string]map[string]struct{}{
+			"sum(up) > 1": {
+				"12:00:00+12:01:00": {},
+				"12:01:00+12:02:00": {},
+				"12:02:00+12:02:30": {},
+			},
+			"max(up) < 1": {
+				"12:00:00+12:01:00": {},
+				"12:01:00+12:02:00": {},
+				"12:02:00+12:02:30": {},
+			},
+		},
+	})
 }

app/vmalert/rule/alerting.go

@@ -183,7 +183,7 @@ func (ar *AlertingRule) GetAlert(id uint64) *notifier.Alert {
 	return ar.alerts[id]
 }
 
-func (ar *AlertingRule) logDebugf(at time.Time, a *notifier.Alert, format string, args ...interface{}) {
+func (ar *AlertingRule) logDebugf(at time.Time, a *notifier.Alert, format string, args ...any) {
 	if !ar.Debug {
 		return
 	}

app/vmalert/rule/group.go

@@ -475,7 +475,7 @@ func delayBeforeStart(ts time.Time, key uint64, interval time.Duration, offset *
 	return randSleep
 }
 
-func (g *Group) infof(format string, args ...interface{}) {
+func (g *Group) infof(format string, args ...any) {
 	msg := fmt.Sprintf(format, args...)
 	logger.Infof("group %q %s; interval=%v; eval_offset=%v; concurrency=%d",
 		g.Name, msg, g.Interval, g.EvalOffset, g.Concurrency)

app/vmalert/rule/group_test.go

@@ -37,158 +37,142 @@ func TestMain(m *testing.M) {
 }
 
 func TestUpdateWith(t *testing.T) {
-	testCases := []struct {
-		name         string
-		currentRules []config.Rule
-		newRules     []config.Rule
-	}{
-		{
-			"new rule",
-			nil,
-			[]config.Rule{{Alert: "bar"}},
-		},
-		{
-			"update alerting rule",
-			[]config.Rule{
-				{
-					Alert: "foo",
-					Expr:  "up > 0",
-					For:   promutils.NewDuration(time.Second),
-					Labels: map[string]string{
-						"bar": "baz",
-					},
-					Annotations: map[string]string{
-						"summary":     "{{ $value|humanize }}",
-						"description": "{{$labels}}",
-					},
-				},
-				{
-					Alert: "bar",
-					Expr:  "up > 0",
-					For:   promutils.NewDuration(time.Second),
-					Labels: map[string]string{
-						"bar": "baz",
-					},
-				},
-			},
-			[]config.Rule{
-				{
-					Alert: "foo",
-					Expr:  "up > 10",
-					For:   promutils.NewDuration(time.Second),
-					Labels: map[string]string{
-						"baz": "bar",
-					},
-					Annotations: map[string]string{
-						"summary": "none",
-					},
-				},
-				{
-					Alert:         "bar",
-					Expr:          "up > 0",
-					For:           promutils.NewDuration(2 * time.Second),
-					KeepFiringFor: promutils.NewDuration(time.Minute),
-					Labels: map[string]string{
-						"bar": "baz",
-					},
-				},
-			},
-		},
-		{
-			"update recording rule",
-			[]config.Rule{{
-				Record: "foo",
-				Expr:   "max(up)",
-				Labels: map[string]string{
-					"bar": "baz",
-				},
-			}},
-			[]config.Rule{{
-				Record: "foo",
-				Expr:   "min(up)",
-				Labels: map[string]string{
-					"baz": "bar",
-				},
-			}},
-		},
-		{
-			"empty rule",
-			[]config.Rule{{Alert: "foo"}, {Record: "bar"}},
-			nil,
-		},
-		{
-			"multiple rules",
-			[]config.Rule{
-				{Alert: "bar"},
-				{Alert: "baz"},
-				{Alert: "foo"},
-			},
-			[]config.Rule{
-				{Alert: "baz"},
-				{Record: "foo"},
-			},
-		},
-		{
-			"replace rule",
-			[]config.Rule{{Alert: "foo1"}},
-			[]config.Rule{{Alert: "foo2"}},
-		},
-		{
-			"replace multiple rules",
-			[]config.Rule{
-				{Alert: "foo1"},
-				{Record: "foo2"},
-				{Alert: "foo3"},
-			},
-			[]config.Rule{
-				{Alert: "foo3"},
-				{Alert: "foo4"},
-				{Record: "foo5"},
-			},
-		},
-	}
+	f := func(currentRules, newRules []config.Rule) {
+		t.Helper()
 
-	for _, tc := range testCases {
-		t.Run(tc.name, func(t *testing.T) {
-			g := &Group{Name: "test"}
-			qb := &datasource.FakeQuerier{}
-			for _, r := range tc.currentRules {
-				r.ID = config.HashRule(r)
-				g.Rules = append(g.Rules, g.newRule(qb, r))
-			}
+		g := &Group{
+			Name: "test",
+		}
+		qb := &datasource.FakeQuerier{}
+		for _, r := range currentRules {
+			r.ID = config.HashRule(r)
+			g.Rules = append(g.Rules, g.newRule(qb, r))
+		}
 
-			ng := &Group{Name: "test"}
-			for _, r := range tc.newRules {
-				r.ID = config.HashRule(r)
-				ng.Rules = append(ng.Rules, ng.newRule(qb, r))
-			}
+		ng := &Group{
+			Name: "test",
+		}
+		for _, r := range newRules {
+			r.ID = config.HashRule(r)
+			ng.Rules = append(ng.Rules, ng.newRule(qb, r))
+		}
 
-			err := g.updateWith(ng)
-			if err != nil {
-				t.Fatal(err)
-			}
+		err := g.updateWith(ng)
+		if err != nil {
+			t.Fatalf("cannot update rule: %s", err)
+		}
 
-			if len(g.Rules) != len(tc.newRules) {
-				t.Fatalf("expected to have %d rules; got: %d",
-					len(g.Rules), len(tc.newRules))
-			}
-			sort.Slice(g.Rules, func(i, j int) bool {
-				return g.Rules[i].ID() < g.Rules[j].ID()
-			})
-			sort.Slice(ng.Rules, func(i, j int) bool {
-				return ng.Rules[i].ID() < ng.Rules[j].ID()
-			})
-			for i, r := range g.Rules {
-				got, want := r, ng.Rules[i]
-				if got.ID() != want.ID() {
-					t.Fatalf("expected to have rule %q; got %q", want, got)
-				}
-				if err := CompareRules(t, got, want); err != nil {
-					t.Fatalf("comparison error: %s", err)
-				}
-			}
+		if len(g.Rules) != len(newRules) {
+			t.Fatalf("expected to have %d rules; got: %d", len(g.Rules), len(newRules))
+		}
+		sort.Slice(g.Rules, func(i, j int) bool {
+			return g.Rules[i].ID() < g.Rules[j].ID()
 		})
+		sort.Slice(ng.Rules, func(i, j int) bool {
+			return ng.Rules[i].ID() < ng.Rules[j].ID()
+		})
+		for i, r := range g.Rules {
+			got, want := r, ng.Rules[i]
+			if got.ID() != want.ID() {
+				t.Fatalf("expected to have rule %q; got %q", want, got)
+			}
+			if err := CompareRules(t, got, want); err != nil {
+				t.Fatalf("comparison error: %s", err)
+			}
+		}
 	}
+
+	// new rule
+	f(nil, []config.Rule{
+		{Alert: "bar"},
+	})
+
+	// update alerting rule
+	f([]config.Rule{
+		{
+			Alert: "foo",
+			Expr:  "up > 0",
+			For:   promutils.NewDuration(time.Second),
+			Labels: map[string]string{
+				"bar": "baz",
+			},
+			Annotations: map[string]string{
+				"summary":     "{{ $value|humanize }}",
+				"description": "{{$labels}}",
+			},
+		},
+		{
+			Alert: "bar",
+			Expr:  "up > 0",
+			For:   promutils.NewDuration(time.Second),
+			Labels: map[string]string{
+				"bar": "baz",
+			},
+		},
+	}, []config.Rule{
+		{
+			Alert: "foo",
+			Expr:  "up > 10",
+			For:   promutils.NewDuration(time.Second),
+			Labels: map[string]string{
+				"baz": "bar",
+			},
+			Annotations: map[string]string{
+				"summary": "none",
+			},
+		},
+		{
+			Alert:         "bar",
+			Expr:          "up > 0",
+			For:           promutils.NewDuration(2 * time.Second),
+			KeepFiringFor: promutils.NewDuration(time.Minute),
+			Labels: map[string]string{
+				"bar": "baz",
+			},
+		},
+	})
+
+	// update recording rule
+	f([]config.Rule{{
+		Record: "foo",
+		Expr:   "max(up)",
+		Labels: map[string]string{
+			"bar": "baz",
+		},
+	}}, []config.Rule{{
+		Record: "foo",
+		Expr:   "min(up)",
+		Labels: map[string]string{
+			"baz": "bar",
+		},
+	}})
+
+	// empty rule
+	f([]config.Rule{{Alert: "foo"}, {Record: "bar"}}, nil)
+
+	// multiple rules
+	f([]config.Rule{
+		{Alert: "bar"},
+		{Alert: "baz"},
+		{Alert: "foo"},
+	}, []config.Rule{
+		{Alert: "baz"},
+		{Record: "foo"},
+	})
+
+	// replace rule
+	f([]config.Rule{{Alert: "foo1"}}, []config.Rule{{Alert: "foo2"}})
+
+	// replace multiple rules
+	f([]config.Rule{
+		{Alert: "foo1"},
+		{Record: "foo2"},
+		{Alert: "foo3"},
+	}, []config.Rule{
+		{Alert: "foo3"},
+		{Alert: "foo4"},
+		{Record: "foo5"},
+	})
 }
 
 func TestGroupStart(t *testing.T) {
@@ -312,30 +296,23 @@ func TestGroupStart(t *testing.T) {
 	<-finished
 }
 
-func TestResolveDuration(t *testing.T) {
-	testCases := []struct {
-		groupInterval time.Duration
-		maxDuration   time.Duration
-		resendDelay   time.Duration
-		expected      time.Duration
-	}{
-		{time.Minute, 0, 0, 4 * time.Minute},
-		{time.Minute, 0, 2 * time.Minute, 8 * time.Minute},
-		{time.Minute, 4 * time.Minute, 4 * time.Minute, 4 * time.Minute},
-		{2 * time.Minute, time.Minute, 2 * time.Minute, time.Minute},
-		{time.Minute, 2 * time.Minute, 1 * time.Minute, 2 * time.Minute},
-		{2 * time.Minute, 0, 1 * time.Minute, 8 * time.Minute},
-		{0, 0, 0, 0},
+func TestGetResolveDuration(t *testing.T) {
+	f := func(groupInterval, maxDuration, resendDelay, resultExpected time.Duration) {
+		t.Helper()
+
+		result := getResolveDuration(groupInterval, resendDelay, maxDuration)
+		if result != resultExpected {
+			t.Fatalf("unexpected result; got %s; want %s", result, resultExpected)
+		}
 	}
 
-	for _, tc := range testCases {
-		t.Run(fmt.Sprintf("%v-%v-%v", tc.groupInterval, tc.expected, tc.maxDuration), func(t *testing.T) {
-			got := getResolveDuration(tc.groupInterval, tc.resendDelay, tc.maxDuration)
-			if got != tc.expected {
-				t.Errorf("expected to have %v; got %v", tc.expected, got)
-			}
-		})
-	}
+	f(0, 0, 0, 0)
+	f(time.Minute, 0, 0, 4*time.Minute)
+	f(time.Minute, 0, 2*time.Minute, 8*time.Minute)
+	f(time.Minute, 4*time.Minute, 4*time.Minute, 4*time.Minute)
+	f(2*time.Minute, time.Minute, 2*time.Minute, time.Minute)
+	f(time.Minute, 2*time.Minute, 1*time.Minute, 2*time.Minute)
+	f(2*time.Minute, 0, 1*time.Minute, 8*time.Minute)
 }
 
 func TestGetStaleSeries(t *testing.T) {
@@ -345,6 +322,7 @@ func TestGetStaleSeries(t *testing.T) {
 	}
 	f := func(r Rule, labels, expLabels [][]prompbmarshal.Label) {
 		t.Helper()
+
 		var tss []prompbmarshal.TimeSeries
 		for _, l := range labels {
 			tss = append(tss, newTimeSeriesPB([]float64{1}, []int64{ts.Unix()}, l))
@@ -606,7 +584,7 @@ func TestGroupStartDelay(t *testing.T) {
 		delay := delayBeforeStart(at, key, g.Interval, g.EvalOffset)
 		gotStart := at.Add(delay)
 		if expTS != gotStart {
-			t.Errorf("expected to get %v; got %v instead", expTS, gotStart)
+			t.Fatalf("expected to get %v; got %v instead", expTS, gotStart)
 		}
 	}
 
@@ -647,157 +625,118 @@ func TestGroupStartDelay(t *testing.T) {
 }
 
 func TestGetPrometheusReqTimestamp(t *testing.T) {
+	f := func(g *Group, tsOrigin, tsExpected string) {
+		t.Helper()
+
+		originT, _ := time.Parse(time.RFC3339, tsOrigin)
+		expT, _ := time.Parse(time.RFC3339, tsExpected)
+		gotTS := g.adjustReqTimestamp(originT)
+		if !gotTS.Equal(expT) {
+			t.Fatalf("get wrong prometheus request timestamp: %s; want %s", gotTS, expT)
+		}
+	}
+
 	offset := 30 * time.Minute
 	evalDelay := 1 * time.Minute
 	disableAlign := false
-	testCases := []struct {
-		name            string
-		g               *Group
-		originTS, expTS string
-	}{
-		{
-			"with query align + default evalDelay",
-			&Group{
-				Interval: time.Hour,
-			},
-			"2023-08-28T11:11:00+00:00",
-			"2023-08-28T11:00:00+00:00",
-		},
-		{
-			"without query align + default evalDelay",
-			&Group{
-				Interval:      time.Hour,
-				evalAlignment: &disableAlign,
-			},
-			"2023-08-28T11:11:00+00:00",
-			"2023-08-28T11:10:30+00:00",
-		},
-		{
-			"with eval_offset, find previous offset point + default evalDelay",
-			&Group{
-				EvalOffset: &offset,
-				Interval:   time.Hour,
-			},
-			"2023-08-28T11:11:00+00:00",
-			"2023-08-28T10:30:00+00:00",
-		},
-		{
-			"with eval_offset + default evalDelay",
-			&Group{
-				EvalOffset: &offset,
-				Interval:   time.Hour,
-			},
-			"2023-08-28T11:41:00+00:00",
-			"2023-08-28T11:30:00+00:00",
-		},
-		{
-			"1h interval with eval_delay",
-			&Group{
-				EvalDelay: &evalDelay,
-				Interval:  time.Hour,
-			},
-			"2023-08-28T11:41:00+00:00",
-			"2023-08-28T11:00:00+00:00",
-		},
-		{
-			"1m interval with eval_delay",
-			&Group{
-				EvalDelay: &evalDelay,
-				Interval:  time.Minute,
-			},
-			"2023-08-28T11:41:13+00:00",
-			"2023-08-28T11:40:00+00:00",
-		},
-		{
-			"disable alignment with eval_delay",
-			&Group{
-				EvalDelay:     &evalDelay,
-				Interval:      time.Hour,
-				evalAlignment: &disableAlign,
-			},
-			"2023-08-28T11:41:00+00:00",
-			"2023-08-28T11:40:00+00:00",
-		},
-	}
-	for _, tc := range testCases {
-		t.Run(tc.name, func(t *testing.T) {
-			originT, _ := time.Parse(time.RFC3339, tc.originTS)
-			expT, _ := time.Parse(time.RFC3339, tc.expTS)
-			gotTS := tc.g.adjustReqTimestamp(originT)
-			if !gotTS.Equal(expT) {
-				t.Fatalf("get wrong prometheus request timestamp, expect %s, got %s", expT, gotTS)
-			}
-		})
-	}
+
+	// with query align + default evalDelay
+	f(&Group{
+		Interval: time.Hour,
+	}, "2023-08-28T11:11:00+00:00", "2023-08-28T11:00:00+00:00")
+
+	// without query align + default evalDelay
+	f(&Group{
+		Interval:      time.Hour,
+		evalAlignment: &disableAlign,
+	}, "2023-08-28T11:11:00+00:00", "2023-08-28T11:10:30+00:00")
+
+	// with eval_offset, find previous offset point + default evalDelay
+	f(&Group{
+		EvalOffset: &offset,
+		Interval:   time.Hour,
+	}, "2023-08-28T11:11:00+00:00", "2023-08-28T10:30:00+00:00")
+
+	// with eval_offset + default evalDelay
+	f(&Group{
+		EvalOffset: &offset,
+		Interval:   time.Hour,
+	}, "2023-08-28T11:41:00+00:00", "2023-08-28T11:30:00+00:00")
+
+	// 1h interval with eval_delay
+	f(&Group{
+		EvalDelay: &evalDelay,
+		Interval:  time.Hour,
+	}, "2023-08-28T11:41:00+00:00", "2023-08-28T11:00:00+00:00")
+
+	// 1m interval with eval_delay
+	f(&Group{
+		EvalDelay: &evalDelay,
+		Interval:  time.Minute,
+	}, "2023-08-28T11:41:13+00:00", "2023-08-28T11:40:00+00:00")
+
+	// disable alignment with eval_delay
+	f(&Group{
+		EvalDelay:     &evalDelay,
+		Interval:      time.Hour,
+		evalAlignment: &disableAlign,
+	}, "2023-08-28T11:41:00+00:00", "2023-08-28T11:40:00+00:00")
 }
 
 func TestRangeIterator(t *testing.T) {
-	testCases := []struct {
-		ri     rangeIterator
-		result [][2]time.Time
-	}{
-		{
-			ri: rangeIterator{
-				start: parseTime(t, "2021-01-01T12:00:00.000Z"),
-				end:   parseTime(t, "2021-01-01T12:30:00.000Z"),
-				step:  5 * time.Minute,
-			},
-			result: [][2]time.Time{
-				{parseTime(t, "2021-01-01T12:00:00.000Z"), parseTime(t, "2021-01-01T12:05:00.000Z")},
-				{parseTime(t, "2021-01-01T12:05:00.000Z"), parseTime(t, "2021-01-01T12:10:00.000Z")},
-				{parseTime(t, "2021-01-01T12:10:00.000Z"), parseTime(t, "2021-01-01T12:15:00.000Z")},
-				{parseTime(t, "2021-01-01T12:15:00.000Z"), parseTime(t, "2021-01-01T12:20:00.000Z")},
-				{parseTime(t, "2021-01-01T12:20:00.000Z"), parseTime(t, "2021-01-01T12:25:00.000Z")},
-				{parseTime(t, "2021-01-01T12:25:00.000Z"), parseTime(t, "2021-01-01T12:30:00.000Z")},
-			},
-		},
-		{
-			ri: rangeIterator{
-				start: parseTime(t, "2021-01-01T12:00:00.000Z"),
-				end:   parseTime(t, "2021-01-01T12:30:00.000Z"),
-				step:  45 * time.Minute,
-			},
-			result: [][2]time.Time{
-				{parseTime(t, "2021-01-01T12:00:00.000Z"), parseTime(t, "2021-01-01T12:30:00.000Z")},
-				{parseTime(t, "2021-01-01T12:30:00.000Z"), parseTime(t, "2021-01-01T12:30:00.000Z")},
-			},
-		},
-		{
-			ri: rangeIterator{
-				start: parseTime(t, "2021-01-01T12:00:12.000Z"),
-				end:   parseTime(t, "2021-01-01T12:00:17.000Z"),
-				step:  time.Second,
-			},
-			result: [][2]time.Time{
-				{parseTime(t, "2021-01-01T12:00:12.000Z"), parseTime(t, "2021-01-01T12:00:13.000Z")},
-				{parseTime(t, "2021-01-01T12:00:13.000Z"), parseTime(t, "2021-01-01T12:00:14.000Z")},
-				{parseTime(t, "2021-01-01T12:00:14.000Z"), parseTime(t, "2021-01-01T12:00:15.000Z")},
-				{parseTime(t, "2021-01-01T12:00:15.000Z"), parseTime(t, "2021-01-01T12:00:16.000Z")},
-				{parseTime(t, "2021-01-01T12:00:16.000Z"), parseTime(t, "2021-01-01T12:00:17.000Z")},
-			},
-		},
+	f := func(ri rangeIterator, resultExpected [][2]time.Time) {
+		t.Helper()
+
+		var j int
+		for ri.next() {
+			if len(resultExpected) < j+1 {
+				t.Fatalf("unexpected result for iterator on step %d: %v - %v", j, ri.s, ri.e)
+			}
+			s, e := ri.s, ri.e
+			expS, expE := resultExpected[j][0], resultExpected[j][1]
+			if s != expS {
+				t.Fatalf("expected to get start=%v; got %v", expS, s)
+			}
+			if e != expE {
+				t.Fatalf("expected to get end=%v; got %v", expE, e)
+			}
+			j++
+		}
 	}
 
-	for i, tc := range testCases {
-		t.Run(fmt.Sprintf("case %d", i), func(t *testing.T) {
-			var j int
-			for tc.ri.next() {
-				if len(tc.result) < j+1 {
-					t.Fatalf("unexpected result for iterator on step %d: %v - %v",
-						j, tc.ri.s, tc.ri.e)
-				}
-				s, e := tc.ri.s, tc.ri.e
-				expS, expE := tc.result[j][0], tc.result[j][1]
-				if s != expS {
-					t.Fatalf("expected to get start=%v; got %v", expS, s)
-				}
-				if e != expE {
-					t.Fatalf("expected to get end=%v; got %v", expE, e)
-				}
-				j++
-			}
-		})
-	}
+	f(rangeIterator{
+		start: parseTime(t, "2021-01-01T12:00:00.000Z"),
+		end:   parseTime(t, "2021-01-01T12:30:00.000Z"),
+		step:  5 * time.Minute,
+	}, [][2]time.Time{
+		{parseTime(t, "2021-01-01T12:00:00.000Z"), parseTime(t, "2021-01-01T12:05:00.000Z")},
+		{parseTime(t, "2021-01-01T12:05:00.000Z"), parseTime(t, "2021-01-01T12:10:00.000Z")},
+		{parseTime(t, "2021-01-01T12:10:00.000Z"), parseTime(t, "2021-01-01T12:15:00.000Z")},
+		{parseTime(t, "2021-01-01T12:15:00.000Z"), parseTime(t, "2021-01-01T12:20:00.000Z")},
+		{parseTime(t, "2021-01-01T12:20:00.000Z"), parseTime(t, "2021-01-01T12:25:00.000Z")},
+		{parseTime(t, "2021-01-01T12:25:00.000Z"), parseTime(t, "2021-01-01T12:30:00.000Z")},
+	})
+
+	f(rangeIterator{
+		start: parseTime(t, "2021-01-01T12:00:00.000Z"),
+		end:   parseTime(t, "2021-01-01T12:30:00.000Z"),
+		step:  45 * time.Minute,
+	}, [][2]time.Time{
+		{parseTime(t, "2021-01-01T12:00:00.000Z"), parseTime(t, "2021-01-01T12:30:00.000Z")},
+		{parseTime(t, "2021-01-01T12:30:00.000Z"), parseTime(t, "2021-01-01T12:30:00.000Z")},
+	})
+
+	f(rangeIterator{
+		start: parseTime(t, "2021-01-01T12:00:12.000Z"),
+		end:   parseTime(t, "2021-01-01T12:00:17.000Z"),
+		step:  time.Second,
+	}, [][2]time.Time{
+		{parseTime(t, "2021-01-01T12:00:12.000Z"), parseTime(t, "2021-01-01T12:00:13.000Z")},
+		{parseTime(t, "2021-01-01T12:00:13.000Z"), parseTime(t, "2021-01-01T12:00:14.000Z")},
+		{parseTime(t, "2021-01-01T12:00:14.000Z"), parseTime(t, "2021-01-01T12:00:15.000Z")},
+		{parseTime(t, "2021-01-01T12:00:15.000Z"), parseTime(t, "2021-01-01T12:00:16.000Z")},
+		{parseTime(t, "2021-01-01T12:00:16.000Z"), parseTime(t, "2021-01-01T12:00:17.000Z")},
+	})
 }
 
 func parseTime(t *testing.T, s string) time.Time {

app/vmalert/rule/recording_test.go

@@ -13,218 +13,225 @@ import (
 )
 
 func TestRecordingRule_Exec(t *testing.T) {
+	f := func(rule *RecordingRule, metrics []datasource.Metric, tssExpected []prompbmarshal.TimeSeries) {
+		t.Helper()
+
+		fq := &datasource.FakeQuerier{}
+		fq.Add(metrics...)
+		rule.q = fq
+		rule.state = &ruleState{
+			entries: make([]StateEntry, 10),
+		}
+		tss, err := rule.exec(context.TODO(), time.Now(), 0)
+		if err != nil {
+			t.Fatalf("unexpected RecordingRule.exec error: %s", err)
+		}
+		if err := compareTimeSeries(t, tssExpected, tss); err != nil {
+			t.Fatalf("timeseries missmatch: %s", err)
+		}
+	}
+
 	timestamp := time.Now()
-	testCases := []struct {
-		rule    *RecordingRule
-		metrics []datasource.Metric
-		expTS   []prompbmarshal.TimeSeries
-	}{
-		{
-			&RecordingRule{Name: "foo"},
-			[]datasource.Metric{metricWithValueAndLabels(t, 10,
-				"__name__", "bar",
-			)},
-			[]prompbmarshal.TimeSeries{
-				newTimeSeries([]float64{10}, []int64{timestamp.UnixNano()}, map[string]string{
-					"__name__": "foo",
-				}),
-			},
+
+	f(&RecordingRule{
+		Name: "foo",
+	}, []datasource.Metric{
+		metricWithValueAndLabels(t, 10, "__name__", "bar"),
+	}, []prompbmarshal.TimeSeries{
+		newTimeSeries([]float64{10}, []int64{timestamp.UnixNano()}, map[string]string{
+			"__name__": "foo",
+		}),
+	})
+
+	f(&RecordingRule{
+		Name: "foobarbaz",
+	}, []datasource.Metric{
+		metricWithValueAndLabels(t, 1, "__name__", "foo", "job", "foo"),
+		metricWithValueAndLabels(t, 2, "__name__", "bar", "job", "bar"),
+		metricWithValueAndLabels(t, 3, "__name__", "baz", "job", "baz"),
+	}, []prompbmarshal.TimeSeries{
+		newTimeSeries([]float64{1}, []int64{timestamp.UnixNano()}, map[string]string{
+			"__name__": "foobarbaz",
+			"job":      "foo",
+		}),
+		newTimeSeries([]float64{2}, []int64{timestamp.UnixNano()}, map[string]string{
+			"__name__": "foobarbaz",
+			"job":      "bar",
+		}),
+		newTimeSeries([]float64{3}, []int64{timestamp.UnixNano()}, map[string]string{
+			"__name__": "foobarbaz",
+			"job":      "baz",
+		}),
+	})
+
+	f(&RecordingRule{
+		Name: "job:foo",
+		Labels: map[string]string{
+			"source": "test",
 		},
-		{
-			&RecordingRule{Name: "foobarbaz"},
-			[]datasource.Metric{
-				metricWithValueAndLabels(t, 1, "__name__", "foo", "job", "foo"),
-				metricWithValueAndLabels(t, 2, "__name__", "bar", "job", "bar"),
-				metricWithValueAndLabels(t, 3, "__name__", "baz", "job", "baz"),
-			},
-			[]prompbmarshal.TimeSeries{
-				newTimeSeries([]float64{1}, []int64{timestamp.UnixNano()}, map[string]string{
-					"__name__": "foobarbaz",
-					"job":      "foo",
-				}),
-				newTimeSeries([]float64{2}, []int64{timestamp.UnixNano()}, map[string]string{
-					"__name__": "foobarbaz",
-					"job":      "bar",
-				}),
-				newTimeSeries([]float64{3}, []int64{timestamp.UnixNano()}, map[string]string{
-					"__name__": "foobarbaz",
-					"job":      "baz",
-				}),
-			},
-		},
-		{
-			&RecordingRule{
-				Name: "job:foo",
-				Labels: map[string]string{
-					"source": "test",
-				},
-			},
-			[]datasource.Metric{
-				metricWithValueAndLabels(t, 2, "__name__", "foo", "job", "foo"),
-				metricWithValueAndLabels(t, 1, "__name__", "bar", "job", "bar", "source", "origin"),
-			},
-			[]prompbmarshal.TimeSeries{
-				newTimeSeries([]float64{2}, []int64{timestamp.UnixNano()}, map[string]string{
-					"__name__": "job:foo",
-					"job":      "foo",
-					"source":   "test",
-				}),
-				newTimeSeries([]float64{1}, []int64{timestamp.UnixNano()}, map[string]string{
-					"__name__":        "job:foo",
-					"job":             "bar",
-					"source":          "test",
-					"exported_source": "origin",
-				}),
-			},
-		},
-	}
-	for _, tc := range testCases {
-		t.Run(tc.rule.Name, func(t *testing.T) {
-			fq := &datasource.FakeQuerier{}
-			fq.Add(tc.metrics...)
-			tc.rule.q = fq
-			tc.rule.state = &ruleState{entries: make([]StateEntry, 10)}
-			tss, err := tc.rule.exec(context.TODO(), time.Now(), 0)
-			if err != nil {
-				t.Fatalf("unexpected Exec err: %s", err)
-			}
-			if err := compareTimeSeries(t, tc.expTS, tss); err != nil {
-				t.Fatalf("timeseries missmatch: %s", err)
-			}
-		})
-	}
+	}, []datasource.Metric{
+		metricWithValueAndLabels(t, 2, "__name__", "foo", "job", "foo"),
+		metricWithValueAndLabels(t, 1, "__name__", "bar", "job", "bar", "source", "origin"),
+	}, []prompbmarshal.TimeSeries{
+		newTimeSeries([]float64{2}, []int64{timestamp.UnixNano()}, map[string]string{
+			"__name__": "job:foo",
+			"job":      "foo",
+			"source":   "test",
+		}),
+		newTimeSeries([]float64{1}, []int64{timestamp.UnixNano()}, map[string]string{
+			"__name__":        "job:foo",
+			"job":             "bar",
+			"source":          "test",
+			"exported_source": "origin",
+		}),
+	})
 }
 
 func TestRecordingRule_ExecRange(t *testing.T) {
-	timestamp := time.Now()
-	testCases := []struct {
-		rule    *RecordingRule
-		metrics []datasource.Metric
-		expTS   []prompbmarshal.TimeSeries
-	}{
-		{
-			&RecordingRule{Name: "foo"},
-			[]datasource.Metric{metricWithValuesAndLabels(t, []float64{10, 20, 30},
-				"__name__", "bar",
-			)},
-			[]prompbmarshal.TimeSeries{
-				newTimeSeries([]float64{10, 20, 30},
-					[]int64{timestamp.UnixNano(), timestamp.UnixNano(), timestamp.UnixNano()},
-					map[string]string{
-						"__name__": "foo",
-					}),
-			},
-		},
-		{
-			&RecordingRule{Name: "foobarbaz"},
-			[]datasource.Metric{
-				metricWithValuesAndLabels(t, []float64{1}, "__name__", "foo", "job", "foo"),
-				metricWithValuesAndLabels(t, []float64{2, 3}, "__name__", "bar", "job", "bar"),
-				metricWithValuesAndLabels(t, []float64{4, 5, 6}, "__name__", "baz", "job", "baz"),
-			},
-			[]prompbmarshal.TimeSeries{
-				newTimeSeries([]float64{1}, []int64{timestamp.UnixNano()}, map[string]string{
-					"__name__": "foobarbaz",
-					"job":      "foo",
-				}),
-				newTimeSeries([]float64{2, 3}, []int64{timestamp.UnixNano(), timestamp.UnixNano()}, map[string]string{
-					"__name__": "foobarbaz",
-					"job":      "bar",
-				}),
-				newTimeSeries([]float64{4, 5, 6},
-					[]int64{timestamp.UnixNano(), timestamp.UnixNano(), timestamp.UnixNano()},
-					map[string]string{
-						"__name__": "foobarbaz",
-						"job":      "baz",
-					}),
-			},
-		},
-		{
-			&RecordingRule{Name: "job:foo", Labels: map[string]string{
-				"source": "test",
-			}},
-			[]datasource.Metric{
-				metricWithValueAndLabels(t, 2, "__name__", "foo", "job", "foo"),
-				metricWithValueAndLabels(t, 1, "__name__", "bar", "job", "bar"),
-			},
-			[]prompbmarshal.TimeSeries{
-				newTimeSeries([]float64{2}, []int64{timestamp.UnixNano()}, map[string]string{
-					"__name__": "job:foo",
-					"job":      "foo",
-					"source":   "test",
-				}),
-				newTimeSeries([]float64{1}, []int64{timestamp.UnixNano()}, map[string]string{
-					"__name__": "job:foo",
-					"job":      "bar",
-					"source":   "test",
-				}),
-			},
-		},
-	}
-	for _, tc := range testCases {
-		t.Run(tc.rule.Name, func(t *testing.T) {
-			fq := &datasource.FakeQuerier{}
-			fq.Add(tc.metrics...)
-			tc.rule.q = fq
-			tss, err := tc.rule.execRange(context.TODO(), time.Now(), time.Now())
-			if err != nil {
-				t.Fatalf("unexpected Exec err: %s", err)
-			}
-			if err := compareTimeSeries(t, tc.expTS, tss); err != nil {
-				t.Fatalf("timeseries missmatch: %s", err)
-			}
-		})
-	}
-}
+	f := func(rule *RecordingRule, metrics []datasource.Metric, tssExpected []prompbmarshal.TimeSeries) {
+		t.Helper()
 
-func TestRecordingRuleLimit(t *testing.T) {
-	timestamp := time.Now()
-	testCases := []struct {
-		limit int
-		err   string
-	}{
-		{
-			limit: 0,
-		},
-		{
-			limit: -1,
-		},
-		{
-			limit: 1,
-			err:   "exec exceeded limit of 1 with 3 series",
-		},
-		{
-			limit: 2,
-			err:   "exec exceeded limit of 2 with 3 series",
-		},
+		fq := &datasource.FakeQuerier{}
+		fq.Add(metrics...)
+		rule.q = fq
+		tss, err := rule.execRange(context.TODO(), time.Now(), time.Now())
+		if err != nil {
+			t.Fatalf("unexpected RecordingRule.execRange error: %s", err)
+		}
+		if err := compareTimeSeries(t, tssExpected, tss); err != nil {
+			t.Fatalf("timeseries missmatch: %s", err)
+		}
 	}
-	testMetrics := []datasource.Metric{
+
+	timestamp := time.Now()
+
+	f(&RecordingRule{
+		Name: "foo",
+	}, []datasource.Metric{
+		metricWithValuesAndLabels(t, []float64{10, 20, 30}, "__name__", "bar"),
+	}, []prompbmarshal.TimeSeries{
+		newTimeSeries([]float64{10, 20, 30}, []int64{timestamp.UnixNano(), timestamp.UnixNano(), timestamp.UnixNano()}, map[string]string{
+			"__name__": "foo",
+		}),
+	})
+
+	f(&RecordingRule{
+		Name: "foobarbaz",
+	}, []datasource.Metric{
 		metricWithValuesAndLabels(t, []float64{1}, "__name__", "foo", "job", "foo"),
 		metricWithValuesAndLabels(t, []float64{2, 3}, "__name__", "bar", "job", "bar"),
 		metricWithValuesAndLabels(t, []float64{4, 5, 6}, "__name__", "baz", "job", "baz"),
-	}
-	rule := &RecordingRule{Name: "job:foo",
-		state: &ruleState{entries: make([]StateEntry, 10)},
+	}, []prompbmarshal.TimeSeries{
+		newTimeSeries([]float64{1}, []int64{timestamp.UnixNano()}, map[string]string{
+			"__name__": "foobarbaz",
+			"job":      "foo",
+		}),
+		newTimeSeries([]float64{2, 3}, []int64{timestamp.UnixNano(), timestamp.UnixNano()}, map[string]string{
+			"__name__": "foobarbaz",
+			"job":      "bar",
+		}),
+		newTimeSeries([]float64{4, 5, 6},
+			[]int64{timestamp.UnixNano(), timestamp.UnixNano(), timestamp.UnixNano()}, map[string]string{
+				"__name__": "foobarbaz",
+				"job":      "baz",
+			}),
+	})
+
+	f(&RecordingRule{
+		Name: "job:foo",
 		Labels: map[string]string{
-			"source": "test_limit",
+			"source": "test",
 		},
-		metrics: &recordingRuleMetrics{
-			errors: utils.GetOrCreateCounter(`vmalert_recording_rules_errors_total{alertname="job:foo"}`),
-		},
-	}
-	var err error
-	for _, testCase := range testCases {
-		fq := &datasource.FakeQuerier{}
-		fq.Add(testMetrics...)
-		rule.q = fq
-		_, err = rule.exec(context.TODO(), timestamp, testCase.limit)
-		if err != nil && !strings.EqualFold(err.Error(), testCase.err) {
-			t.Fatal(err)
-		}
-	}
+	}, []datasource.Metric{
+		metricWithValueAndLabels(t, 2, "__name__", "foo", "job", "foo"),
+		metricWithValueAndLabels(t, 1, "__name__", "bar", "job", "bar"),
+	}, []prompbmarshal.TimeSeries{
+		newTimeSeries([]float64{2}, []int64{timestamp.UnixNano()}, map[string]string{
+			"__name__": "job:foo",
+			"job":      "foo",
+			"source":   "test",
+		}),
+		newTimeSeries([]float64{1}, []int64{timestamp.UnixNano()}, map[string]string{
+			"__name__": "job:foo",
+			"job":      "bar",
+			"source":   "test",
+		}),
+	})
 }
 
-func TestRecordingRule_ExecNegative(t *testing.T) {
+func TestRecordingRuleLimit_Failure(t *testing.T) {
+	f := func(limit int, errStrExpected string) {
+		t.Helper()
+
+		testMetrics := []datasource.Metric{
+			metricWithValuesAndLabels(t, []float64{1}, "__name__", "foo", "job", "foo"),
+			metricWithValuesAndLabels(t, []float64{2, 3}, "__name__", "bar", "job", "bar"),
+			metricWithValuesAndLabels(t, []float64{4, 5, 6}, "__name__", "baz", "job", "baz"),
+		}
+
+		fq := &datasource.FakeQuerier{}
+		fq.Add(testMetrics...)
+
+		rule := &RecordingRule{Name: "job:foo",
+			state: &ruleState{entries: make([]StateEntry, 10)},
+			Labels: map[string]string{
+				"source": "test_limit",
+			},
+			metrics: &recordingRuleMetrics{
+				errors: utils.GetOrCreateCounter(`vmalert_recording_rules_errors_total{alertname="job:foo"}`),
+			},
+		}
+		rule.q = fq
+
+		_, err := rule.exec(context.TODO(), time.Now(), limit)
+		if err == nil {
+			t.Fatalf("expecting non-nil error")
+		}
+		errStr := err.Error()
+		if !strings.Contains(errStr, errStrExpected) {
+			t.Fatalf("missing %q in the error %q", errStrExpected, errStr)
+		}
+	}
+
+	f(1, "exec exceeded limit of 1 with 3 series")
+	f(2, "exec exceeded limit of 2 with 3 series")
+}
+
+func TestRecordingRuleLimit_Success(t *testing.T) {
+	f := func(limit int) {
+		t.Helper()
+
+		testMetrics := []datasource.Metric{
+			metricWithValuesAndLabels(t, []float64{1}, "__name__", "foo", "job", "foo"),
+			metricWithValuesAndLabels(t, []float64{2, 3}, "__name__", "bar", "job", "bar"),
+			metricWithValuesAndLabels(t, []float64{4, 5, 6}, "__name__", "baz", "job", "baz"),
+		}
+
+		fq := &datasource.FakeQuerier{}
+		fq.Add(testMetrics...)
+
+		rule := &RecordingRule{Name: "job:foo",
+			state: &ruleState{entries: make([]StateEntry, 10)},
+			Labels: map[string]string{
+				"source": "test_limit",
+			},
+			metrics: &recordingRuleMetrics{
+				errors: utils.GetOrCreateCounter(`vmalert_recording_rules_errors_total{alertname="job:foo"}`),
+			},
+		}
+		rule.q = fq
+
+		_, err := rule.exec(context.TODO(), time.Now(), limit)
+		if err != nil {
+			t.Fatalf("unexpected error: %s", err)
+		}
+	}
+
+	f(0)
+	f(-1)
+}
+
+func TestRecordingRuleExec_Negative(t *testing.T) {
 	rr := &RecordingRule{
 		Name: "job:foo",
 		Labels: map[string]string{
@@ -256,6 +263,6 @@ func TestRecordingRule_ExecNegative(t *testing.T) {
 
 	_, err = rr.exec(context.TODO(), time.Now(), 0)
 	if err != nil {
-		t.Fatal(err)
+		t.Fatalf("cannot execute recroding rule: %s", err)
 	}
 }

app/vmalert/templates/template.go

@@ -316,7 +316,7 @@ func templateFuncs() textTpl.FuncMap {
 
 		// humanize converts given number to a human readable format
 		// by adding metric prefixes https://en.wikipedia.org/wiki/Metric_prefix
-		"humanize": func(i interface{}) (string, error) {
+		"humanize": func(i any) (string, error) {
 			v, err := toFloat64(i)
 			if err != nil {
 				return "", err
@@ -347,7 +347,7 @@ func templateFuncs() textTpl.FuncMap {
 		},
 
 		// humanize1024 converts given number to a human readable format with 1024 as base
-		"humanize1024": func(i interface{}) (string, error) {
+		"humanize1024": func(i any) (string, error) {
 			v, err := toFloat64(i)
 			if err != nil {
 				return "", err
@@ -359,7 +359,7 @@ func templateFuncs() textTpl.FuncMap {
 		},
 
 		// humanizeDuration converts given seconds to a human-readable duration
-		"humanizeDuration": func(i interface{}) (string, error) {
+		"humanizeDuration": func(i any) (string, error) {
 			v, err := toFloat64(i)
 			if err != nil {
 				return "", err
@@ -405,7 +405,7 @@ func templateFuncs() textTpl.FuncMap {
 		},
 
 		// humanizePercentage converts given ratio value to a fraction of 100
-		"humanizePercentage": func(i interface{}) (string, error) {
+		"humanizePercentage": func(i any) (string, error) {
 			v, err := toFloat64(i)
 			if err != nil {
 				return "", err
@@ -414,7 +414,7 @@ func templateFuncs() textTpl.FuncMap {
 		},
 
 		// humanizeTimestamp converts given timestamp to a human readable time equivalent
-		"humanizeTimestamp": func(i interface{}) (string, error) {
+		"humanizeTimestamp": func(i any) (string, error) {
 			v, err := toFloat64(i)
 			if err != nil {
 				return "", err
@@ -427,7 +427,7 @@ func templateFuncs() textTpl.FuncMap {
 		},
 
 		// toTime converts given timestamp to a time.Time.
-		"toTime": func(i interface{}) (time.Time, error) {
+		"toTime": func(i any) (time.Time, error) {
 			v, err := toFloat64(i)
 			if err != nil {
 				return time.Time{}, err
@@ -524,8 +524,8 @@ func templateFuncs() textTpl.FuncMap {
 
 		// Converts a list of objects to a map with keys arg0, arg1 etc.
 		// This is intended to allow multiple arguments to be passed to templates.
-		"args": func(args ...interface{}) map[string]interface{} {
-			result := make(map[string]interface{})
+		"args": func(args ...any) map[string]any {
+			result := make(map[string]any)
 			for i, a := range args {
 				result[fmt.Sprintf("arg%d", i)] = a
 			}
@@ -565,7 +565,7 @@ func (t Time) Time() time.Time {
 	return time.Unix(int64(t)/second, (int64(t)%second)*nanosPerTick)
 }
 
-func toFloat64(v interface{}) (float64, error) {
+func toFloat64(v any) (float64, error) {
 	switch i := v.(type) {
 	case float64:
 		return i, nil

app/vmalert/templates/template_test.go

@@ -7,10 +7,11 @@ import (
 	textTpl "text/template"
 )
 
-func TestTemplateFuncs(t *testing.T) {
-	funcs := templateFuncs()
+func TestTemplateFuncs_StringConversion(t *testing.T) {
 	f := func(funcName, s, resultExpected string) {
 		t.Helper()
+
+		funcs := templateFuncs()
 		v := funcs[funcName]
 		fLocal := v.(func(s string) string)
 		result := fLocal(s)
@@ -18,6 +19,7 @@ func TestTemplateFuncs(t *testing.T) {
 			t.Fatalf("unexpected result for %s(%q); got\n%s\nwant\n%s", funcName, s, result, resultExpected)
 		}
 	}
+
 	f("title", "foo bar", "Foo Bar")
 	f("toUpper", "foo", "FOO")
 	f("toLower", "FOO", "foo")
@@ -31,7 +33,10 @@ func TestTemplateFuncs(t *testing.T) {
 	f("stripPort", "foo:1234", "foo")
 	f("stripDomain", "foo.bar.baz", "foo")
 	f("stripDomain", "foo.bar:123", "foo:123")
+}
 
+func TestTemplateFuncs_Match(t *testing.T) {
+	funcs := templateFuncs()
 	// check "match" func
 	matchFunc := funcs["match"].(func(pattern, s string) (bool, error))
 	if _, err := matchFunc("invalid[regexp", "abc"); err == nil {
@@ -51,11 +56,15 @@ func TestTemplateFuncs(t *testing.T) {
 	if !ok {
 		t.Fatalf("unexpected mismatch")
 	}
+}
 
-	formatting := func(funcName string, p interface{}, resultExpected string) {
+func TestTemplateFuncs_Formatting(t *testing.T) {
+	f := func(funcName string, p any, resultExpected string) {
 		t.Helper()
+
+		funcs := templateFuncs()
 		v := funcs[funcName]
-		fLocal := v.(func(s interface{}) (string, error))
+		fLocal := v.(func(s any) (string, error))
 		result, err := fLocal(p)
 		if err != nil {
 			t.Fatalf("unexpected error for %s(%f): %s", funcName, p, err)
@@ -64,35 +73,36 @@ func TestTemplateFuncs(t *testing.T) {
 			t.Fatalf("unexpected result for %s(%f); got\n%s\nwant\n%s", funcName, p, result, resultExpected)
 		}
 	}
-	formatting("humanize1024", float64(0), "0")
-	formatting("humanize1024", math.Inf(0), "+Inf")
-	formatting("humanize1024", math.NaN(), "NaN")
-	formatting("humanize1024", float64(127087), "124.1ki")
-	formatting("humanize1024", float64(130137088), "124.1Mi")
-	formatting("humanize1024", float64(133260378112), "124.1Gi")
-	formatting("humanize1024", float64(136458627186688), "124.1Ti")
-	formatting("humanize1024", float64(139733634239168512), "124.1Pi")
-	formatting("humanize1024", float64(143087241460908556288), "124.1Ei")
-	formatting("humanize1024", float64(146521335255970361638912), "124.1Zi")
-	formatting("humanize1024", float64(150037847302113650318245888), "124.1Yi")
-	formatting("humanize1024", float64(153638755637364377925883789312), "1.271e+05Yi")
 
-	formatting("humanize", float64(127087), "127.1k")
-	formatting("humanize", float64(136458627186688), "136.5T")
+	f("humanize1024", float64(0), "0")
+	f("humanize1024", math.Inf(0), "+Inf")
+	f("humanize1024", math.NaN(), "NaN")
+	f("humanize1024", float64(127087), "124.1ki")
+	f("humanize1024", float64(130137088), "124.1Mi")
+	f("humanize1024", float64(133260378112), "124.1Gi")
+	f("humanize1024", float64(136458627186688), "124.1Ti")
+	f("humanize1024", float64(139733634239168512), "124.1Pi")
+	f("humanize1024", float64(143087241460908556288), "124.1Ei")
+	f("humanize1024", float64(146521335255970361638912), "124.1Zi")
+	f("humanize1024", float64(150037847302113650318245888), "124.1Yi")
+	f("humanize1024", float64(153638755637364377925883789312), "1.271e+05Yi")
 
-	formatting("humanizeDuration", 1, "1s")
-	formatting("humanizeDuration", 0.2, "200ms")
-	formatting("humanizeDuration", 42000, "11h 40m 0s")
-	formatting("humanizeDuration", 16790555, "194d 8h 2m 35s")
+	f("humanize", float64(127087), "127.1k")
+	f("humanize", float64(136458627186688), "136.5T")
 
-	formatting("humanizePercentage", 1, "100%")
-	formatting("humanizePercentage", 0.8, "80%")
-	formatting("humanizePercentage", 0.015, "1.5%")
+	f("humanizeDuration", 1, "1s")
+	f("humanizeDuration", 0.2, "200ms")
+	f("humanizeDuration", 42000, "11h 40m 0s")
+	f("humanizeDuration", 16790555, "194d 8h 2m 35s")
 
-	formatting("humanizeTimestamp", 1679055557, "2023-03-17 12:19:17 +0000 UTC")
+	f("humanizePercentage", 1, "100%")
+	f("humanizePercentage", 0.8, "80%")
+	f("humanizePercentage", 0.015, "1.5%")
+
+	f("humanizeTimestamp", 1679055557, "2023-03-17 12:19:17 +0000 UTC")
 }
 
-func mkTemplate(current, replacement interface{}) textTemplate {
+func mkTemplate(current, replacement any) textTemplate {
 	tmpl := textTemplate{}
 	if current != nil {
 		switch val := current.(type) {
@@ -138,224 +148,201 @@ func equalTemplates(tmpls ...*textTpl.Template) bool {
 	return true
 }
 
-func TestTemplates_Load(t *testing.T) {
-	testCases := []struct {
-		name             string
-		initialTemplate  textTemplate
-		pathPatterns     []string
-		overwrite        bool
-		expectedTemplate textTemplate
-		expErr           string
-	}{
-		{
-			"non existing path undefined template override",
-			mkTemplate(nil, nil),
-			[]string{
-				"templates/non-existing/good-*.tpl",
-				"templates/absent/good-*.tpl",
-			},
-			true,
-			mkTemplate(``, nil),
-			"",
-		},
-		{
-			"non existing path defined template override",
-			mkTemplate(`
-				{{- define "test.1" -}}
-					{{- printf "value" -}}
-				{{- end -}}
-			`, nil),
-			[]string{
-				"templates/non-existing/good-*.tpl",
-				"templates/absent/good-*.tpl",
-			},
-			true,
-			mkTemplate(``, nil),
-			"",
-		},
-		{
-			"existing path undefined template override",
-			mkTemplate(nil, nil),
-			[]string{
-				"templates/other/nested/good0-*.tpl",
-				"templates/test/good0-*.tpl",
-			},
-			false,
-			mkTemplate(`
-				{{- define "good0-test.tpl" -}}{{- end -}}
-				{{- define "test.0" -}}
-					{{ printf "Hello %s!" externalURL }}
-				{{- end -}}
-				{{- define "test.1" -}}
-					{{ printf "Hello %s!" externalURL }}
-				{{- end -}}
-				{{- define "test.2" -}}
-					{{ printf "Hello %s!" externalURL }}
-				{{- end -}}
-				{{- define "test.3" -}}
-					{{ printf "Hello %s!" externalURL }}
-				{{- end -}}
-			`, nil),
-			"",
-		},
-		{
-			"existing path defined template override",
-			mkTemplate(`
-				{{- define "test.1" -}}
-					{{ printf "Hello %s!" "world" }}
-				{{- end -}}
-			`, nil),
-			[]string{
-				"templates/other/nested/good0-*.tpl",
-				"templates/test/good0-*.tpl",
-			},
-			false,
-			mkTemplate(`
-				{{- define "good0-test.tpl" -}}{{- end -}}
-				{{- define "test.0" -}}
-					{{ printf "Hello %s!" externalURL }}
-				{{- end -}}
-				{{- define "test.1" -}}
-					{{ printf "Hello %s!" "world" }}
-				{{- end -}}
-				{{- define "test.2" -}}
-					{{ printf "Hello %s!" externalURL }}
-				{{- end -}}
-				{{- define "test.3" -}}
-					{{ printf "Hello %s!" externalURL }}
-				{{- end -}}
-				`, `
-				{{- define "good0-test.tpl" -}}{{- end -}}
-				{{- define "test.0" -}}
-					{{ printf "Hello %s!" externalURL }}
-				{{- end -}}
-				{{- define "test.1" -}}
-					{{ printf "Hello %s!" externalURL }}
-				{{- end -}}
-				{{- define "test.2" -}}
-					{{ printf "Hello %s!" externalURL }}
-				{{- end -}}
-				{{- define "test.3" -}}
-					{{ printf "Hello %s!" externalURL }}
-				{{- end -}}
-			`),
-			"",
-		},
-		{
-			"load template with syntax error",
-			mkTemplate(`
-				{{- define "test.1" -}}
-					{{ printf "Hello %s!" "world" }}
-				{{- end -}}
-			`, nil),
-			[]string{
-				"templates/other/nested/bad0-*.tpl",
-				"templates/test/good0-*.tpl",
-			},
-			false,
-			mkTemplate(`
-				{{- define "test.1" -}}
-					{{ printf "Hello %s!" "world" }}
-				{{- end -}}
-			`, nil),
-			"failed to parse template glob",
-		},
+func TestTemplatesLoad_Failure(t *testing.T) {
+	f := func(pathPatterns []string, expectedErrStr string) {
+		t.Helper()
+
+		err := Load(pathPatterns, false)
+		if err == nil {
+			t.Fatalf("expecting non-nil error")
+		}
+
+		errStr := err.Error()
+		if !strings.Contains(errStr, expectedErrStr) {
+			t.Fatalf("the returned error %q doesn't contain %q", errStr, expectedErrStr)
+		}
 	}
 
-	for _, tc := range testCases {
-		t.Run(tc.name, func(t *testing.T) {
-			masterTmpl = tc.initialTemplate
-			err := Load(tc.pathPatterns, tc.overwrite)
-			if tc.expErr == "" && err != nil {
-				t.Error("happened error that wasn't expected: %w", err)
-			}
-			if tc.expErr != "" && err == nil {
-				t.Error("%+w", err)
-				t.Error("expected error that didn't happened")
-			}
-			if err != nil && !strings.Contains(err.Error(), tc.expErr) {
-				t.Error("%+w", err)
-				t.Error("expected string doesn't exist in error message")
-			}
-			if !equalTemplates(masterTmpl.replacement, tc.expectedTemplate.replacement) {
-				t.Fatalf("replacement template is not as expected")
-			}
-			if !equalTemplates(masterTmpl.current, tc.expectedTemplate.current) {
-				t.Fatalf("current template is not as expected")
-			}
-		})
-	}
+	// load template with syntax error
+	f([]string{
+		"templates/other/nested/bad0-*.tpl",
+		"templates/test/good0-*.tpl",
+	}, "failed to parse template glob")
 }
 
-func TestTemplates_Reload(t *testing.T) {
-	testCases := []struct {
-		name             string
-		initialTemplate  textTemplate
-		expectedTemplate textTemplate
-	}{
-		{
-			"empty current and replacement templates",
-			mkTemplate(nil, nil),
-			mkTemplate(nil, nil),
-		},
-		{
-			"empty current template only",
-			mkTemplate(`
-				{{- define "test.1" -}}
-					{{- printf "value" -}}
-				{{- end -}}
-			`, nil),
-			mkTemplate(`
-				{{- define "test.1" -}}
-					{{- printf "value" -}}
-				{{- end -}}
-			`, nil),
-		},
-		{
-			"empty replacement template only",
-			mkTemplate(nil, `
-				{{- define "test.1" -}}
-					{{- printf "value" -}}
-				{{- end -}}
-			`),
-			mkTemplate(`
-				{{- define "test.1" -}}
-					{{- printf "value" -}}
-				{{- end -}}
-			`, nil),
-		},
-		{
-			"defined both templates",
-			mkTemplate(`
-				{{- define "test.0" -}}
-					{{- printf "value" -}}
-				{{- end -}}
-				{{- define "test.1" -}}
-					{{- printf "before" -}}
-				{{- end -}}
-			`, `
-				{{- define "test.1" -}}
-					{{- printf "after" -}}
-				{{- end -}}
-			`),
-			mkTemplate(`
-				{{- define "test.1" -}}
-					{{- printf "after" -}}
-				{{- end -}}
-			`, nil),
-		},
+func TestTemplatesLoad_Success(t *testing.T) {
+	f := func(initialTmpl textTemplate, pathPatterns []string, overwrite bool, expectedTmpl textTemplate) {
+		t.Helper()
+
+		masterTmplOrig := masterTmpl
+		masterTmpl = initialTmpl
+		defer func() {
+			masterTmpl = masterTmplOrig
+		}()
+
+		if err := Load(pathPatterns, overwrite); err != nil {
+			t.Fatalf("cannot load templates: %s", err)
+		}
+
+		if !equalTemplates(masterTmpl.replacement, expectedTmpl.replacement) {
+			t.Fatalf("unexpected replacement template\ngot\n%+v\nwant\n%+v", masterTmpl.replacement, expectedTmpl.replacement)
+		}
+		if !equalTemplates(masterTmpl.current, expectedTmpl.current) {
+			t.Fatalf("unexpected current template\ngot\n%+v\nwant\n%+v", masterTmpl.current, expectedTmpl.current)
+		}
 	}
 
-	for _, tc := range testCases {
-		t.Run(tc.name, func(t *testing.T) {
-			masterTmpl = tc.initialTemplate
-			Reload()
-			if !equalTemplates(masterTmpl.replacement, tc.expectedTemplate.replacement) {
-				t.Fatalf("replacement template is not as expected")
-			}
-			if !equalTemplates(masterTmpl.current, tc.expectedTemplate.current) {
-				t.Fatalf("current template is not as expected")
-			}
-		})
+	// non existing path undefined template override
+	initialTmpl := mkTemplate(nil, nil)
+	pathPatterns := []string{
+		"templates/non-existing/good-*.tpl",
+		"templates/absent/good-*.tpl",
 	}
+	overwrite := true
+	expectedTmpl := mkTemplate(``, nil)
+	f(initialTmpl, pathPatterns, overwrite, expectedTmpl)
+
+	// non existing path defined template override
+	initialTmpl = mkTemplate(`
+		{{- define "test.1" -}}
+			{{- printf "value" -}}
+		{{- end -}}
+	`, nil)
+	pathPatterns = []string{
+		"templates/non-existing/good-*.tpl",
+		"templates/absent/good-*.tpl",
+	}
+	overwrite = true
+	expectedTmpl = mkTemplate(``, nil)
+	f(initialTmpl, pathPatterns, overwrite, expectedTmpl)
+
+	// existing path undefined template override
+	initialTmpl = mkTemplate(nil, nil)
+	pathPatterns = []string{
+		"templates/other/nested/good0-*.tpl",
+		"templates/test/good0-*.tpl",
+	}
+	overwrite = false
+	expectedTmpl = mkTemplate(`
+		{{- define "good0-test.tpl" -}}{{- end -}}
+		{{- define "test.0" -}}
+			{{ printf "Hello %s!" externalURL }}
+		{{- end -}}
+		{{- define "test.1" -}}
+			{{ printf "Hello %s!" externalURL }}
+		{{- end -}}
+		{{- define "test.2" -}}
+			{{ printf "Hello %s!" externalURL }}
+		{{- end -}}
+		{{- define "test.3" -}}
+			{{ printf "Hello %s!" externalURL }}
+		{{- end -}}
+	`, nil)
+	f(initialTmpl, pathPatterns, overwrite, expectedTmpl)
+
+	// existing path defined template override
+	initialTmpl = mkTemplate(`
+		{{- define "test.1" -}}
+			{{ printf "Hello %s!" "world" }}
+		{{- end -}}
+	`, nil)
+	pathPatterns = []string{
+		"templates/other/nested/good0-*.tpl",
+		"templates/test/good0-*.tpl",
+	}
+	overwrite = false
+	expectedTmpl = mkTemplate(`
+		{{- define "good0-test.tpl" -}}{{- end -}}
+		{{- define "test.0" -}}
+			{{ printf "Hello %s!" externalURL }}
+		{{- end -}}
+		{{- define "test.1" -}}
+			{{ printf "Hello %s!" "world" }}
+		{{- end -}}
+		{{- define "test.2" -}}
+			{{ printf "Hello %s!" externalURL }}
+		{{- end -}}
+		{{- define "test.3" -}}
+			{{ printf "Hello %s!" externalURL }}
+		{{- end -}}
+	`, `
+		{{- define "good0-test.tpl" -}}{{- end -}}
+		{{- define "test.0" -}}
+			{{ printf "Hello %s!" externalURL }}
+		{{- end -}}
+		{{- define "test.1" -}}
+			{{ printf "Hello %s!" externalURL }}
+		{{- end -}}
+		{{- define "test.2" -}}
+			{{ printf "Hello %s!" externalURL }}
+		{{- end -}}
+		{{- define "test.3" -}}
+			{{ printf "Hello %s!" externalURL }}
+		{{- end -}}
+	`)
+	f(initialTmpl, pathPatterns, overwrite, expectedTmpl)
+}
+
+func TestTemplatesReload(t *testing.T) {
+	f := func(initialTmpl, expectedTmpl textTemplate) {
+		t.Helper()
+
+		masterTmplOrig := masterTmpl
+		masterTmpl = initialTmpl
+		defer func() {
+			masterTmpl = masterTmplOrig
+		}()
+
+		Reload()
+
+		if !equalTemplates(masterTmpl.replacement, expectedTmpl.replacement) {
+			t.Fatalf("unexpected replacement template\ngot\n%+v\nwant\n%+v", masterTmpl.replacement, expectedTmpl.replacement)
+		}
+		if !equalTemplates(masterTmpl.current, expectedTmpl.current) {
+			t.Fatalf("unexpected current template\ngot\n%+v\nwant\n%+v", masterTmpl.current, expectedTmpl.current)
+		}
+	}
+
+	// empty current and replacement templates
+	f(mkTemplate(nil, nil), mkTemplate(nil, nil))
+
+	// empty current template only
+	f(mkTemplate(`
+		{{- define "test.1" -}}
+			{{- printf "value" -}}
+		{{- end -}}
+	`, nil), mkTemplate(`
+		{{- define "test.1" -}}
+			{{- printf "value" -}}
+		{{- end -}}
+	`, nil))
+
+	// empty replacement template only
+	f(mkTemplate(nil, `
+		{{- define "test.1" -}}
+			{{- printf "value" -}}
+		{{- end -}}
+	`), mkTemplate(`
+		{{- define "test.1" -}}
+			{{- printf "value" -}}
+		{{- end -}}
+	`, nil))
+
+	// defined both templates
+	f(mkTemplate(`
+		{{- define "test.0" -}}
+			{{- printf "value" -}}
+		{{- end -}}
+		{{- define "test.1" -}}
+			{{- printf "before" -}}
+		{{- end -}}
+	`, `
+		{{- define "test.1" -}}
+			{{- printf "after" -}}
+		{{- end -}}
+	`), mkTemplate(`
+		{{- define "test.1" -}}
+			{{- printf "after" -}}
+		{{- end -}}
+	`, nil))
 }

app/vmalert/utils/err_group_test.go

@@ -7,35 +7,31 @@ import (
 )
 
 func TestErrGroup(t *testing.T) {
-	testCases := []struct {
-		errs []error
-		exp  string
-	}{
-		{nil, ""},
-		{[]error{errors.New("timeout")}, "errors(1): timeout"},
-		{
-			[]error{errors.New("timeout"), errors.New("deadline")},
-			"errors(2): timeout\ndeadline",
-		},
-	}
-	for _, tc := range testCases {
-		eg := new(ErrGroup)
-		for _, err := range tc.errs {
+	f := func(errs []error, resultExpected string) {
+		t.Helper()
+
+		eg := &ErrGroup{}
+		for _, err := range errs {
 			eg.Add(err)
 		}
-		if len(tc.errs) == 0 {
+		if len(errs) == 0 {
 			if eg.Err() != nil {
 				t.Fatalf("expected to get nil error")
 			}
-			continue
+			return
 		}
 		if eg.Err() == nil {
 			t.Fatalf("expected to get non-nil error")
 		}
-		if eg.Error() != tc.exp {
-			t.Fatalf("expected to have: \n%q\ngot:\n%q", tc.exp, eg.Error())
+		result := eg.Error()
+		if result != resultExpected {
+			t.Fatalf("unexpected result\ngot\n%v\nwant\n%v", result, resultExpected)
 		}
 	}
+
+	f(nil, "")
+	f([]error{errors.New("timeout")}, "errors(1): timeout")
+	f([]error{errors.New("timeout"), errors.New("deadline")}, "errors(2): timeout\ndeadline")
 }
 
 // TestErrGroupConcurrent supposed to test concurrent

app/vmalert/web.go

@@ -19,7 +19,7 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/procutil"
 )
 
-var reloadAuthKey = flagutil.NewPassword("reloadAuthKey", "Auth key for /-/reload http endpoint. It must be passed via authKey query arg. It overrides httpAuth.* settings.")
+var reloadAuthKey = flagutil.NewPassword("reloadAuthKey", "Auth key for /-/reload http endpoint. It must be passed via authKey query arg. It overrides -httpAuth.*")
 
 var (
 	apiLinks = [][2]string{
@@ -35,7 +35,7 @@ var (
 		{"-/reload", "reload configuration"},
 	}
 	navItems = []tpl.NavItem{
-		{Name: "vmalert", Url: "."},
+		{Name: "vmalert", Url: "../vmalert"},
 		{Name: "Groups", Url: "groups"},
 		{Name: "Alerts", Url: "alerts"},
 		{Name: "Notifiers", Url: "notifiers"},
@@ -167,7 +167,7 @@ func (rh *requestHandler) handler(w http.ResponseWriter, r *http.Request) bool {
 		w.Write(data)
 		return true
 	case "/-/reload":
-		if !httpserver.CheckAuthFlag(w, r, reloadAuthKey.Get(), "reloadAuthKey") {
+		if !httpserver.CheckAuthFlag(w, r, reloadAuthKey) {
 			return true
 		}
 		logger.Infof("api config reload was called, sending sighup")

app/vmalert/web_test.go

@@ -36,23 +36,23 @@ func TestHandler(t *testing.T) {
 	}}
 	rh := &requestHandler{m: m}
 
-	getResp := func(t *testing.T, url string, to interface{}, code int) {
+	getResp := func(t *testing.T, url string, to any, code int) {
 		t.Helper()
 		resp, err := http.Get(url)
 		if err != nil {
 			t.Fatalf("unexpected err %s", err)
 		}
 		if code != resp.StatusCode {
-			t.Errorf("unexpected status code %d want %d", resp.StatusCode, code)
+			t.Fatalf("unexpected status code %d want %d", resp.StatusCode, code)
 		}
 		defer func() {
 			if err := resp.Body.Close(); err != nil {
-				t.Errorf("err closing body %s", err)
+				t.Fatalf("err closing body %s", err)
 			}
 		}()
 		if to != nil {
 			if err = json.NewDecoder(resp.Body).Decode(to); err != nil {
-				t.Errorf("unexpected err %s", err)
+				t.Fatalf("unexpected err %s", err)
 			}
 		}
 	}
@@ -92,13 +92,13 @@ func TestHandler(t *testing.T) {
 		lr := listAlertsResponse{}
 		getResp(t, ts.URL+"/api/v1/alerts", &lr, 200)
 		if length := len(lr.Data.Alerts); length != 1 {
-			t.Errorf("expected 1 alert got %d", length)
+			t.Fatalf("expected 1 alert got %d", length)
 		}
 
 		lr = listAlertsResponse{}
 		getResp(t, ts.URL+"/vmalert/api/v1/alerts", &lr, 200)
 		if length := len(lr.Data.Alerts); length != 1 {
-			t.Errorf("expected 1 alert got %d", length)
+			t.Fatalf("expected 1 alert got %d", length)
 		}
 	})
 	t.Run("/api/v1/alert?alertID&groupID", func(t *testing.T) {
@@ -106,13 +106,13 @@ func TestHandler(t *testing.T) {
 		alert := &apiAlert{}
 		getResp(t, ts.URL+"/"+expAlert.APILink(), alert, 200)
 		if !reflect.DeepEqual(alert, expAlert) {
-			t.Errorf("expected %v is equal to %v", alert, expAlert)
+			t.Fatalf("expected %v is equal to %v", alert, expAlert)
 		}
 
 		alert = &apiAlert{}
 		getResp(t, ts.URL+"/vmalert/"+expAlert.APILink(), alert, 200)
 		if !reflect.DeepEqual(alert, expAlert) {
-			t.Errorf("expected %v is equal to %v", alert, expAlert)
+			t.Fatalf("expected %v is equal to %v", alert, expAlert)
 		}
 	})
 
@@ -135,13 +135,13 @@ func TestHandler(t *testing.T) {
 		lr := listGroupsResponse{}
 		getResp(t, ts.URL+"/api/v1/rules", &lr, 200)
 		if length := len(lr.Data.Groups); length != 1 {
-			t.Errorf("expected 1 group got %d", length)
+			t.Fatalf("expected 1 group got %d", length)
 		}
 
 		lr = listGroupsResponse{}
 		getResp(t, ts.URL+"/vmalert/api/v1/rules", &lr, 200)
 		if length := len(lr.Data.Groups); length != 1 {
-			t.Errorf("expected 1 group got %d", length)
+			t.Fatalf("expected 1 group got %d", length)
 		}
 	})
 	t.Run("/api/v1/rule?ruleID&groupID", func(t *testing.T) {
@@ -150,14 +150,14 @@ func TestHandler(t *testing.T) {
 		getResp(t, ts.URL+"/"+expRule.APILink(), &gotRule, 200)
 
 		if expRule.ID != gotRule.ID {
-			t.Errorf("expected to get Rule %q; got %q instead", expRule.ID, gotRule.ID)
+			t.Fatalf("expected to get Rule %q; got %q instead", expRule.ID, gotRule.ID)
 		}
 
 		gotRule = apiRule{}
 		getResp(t, ts.URL+"/vmalert/"+expRule.APILink(), &gotRule, 200)
 
 		if expRule.ID != gotRule.ID {
-			t.Errorf("expected to get Rule %q; got %q instead", expRule.ID, gotRule.ID)
+			t.Fatalf("expected to get Rule %q; got %q instead", expRule.ID, gotRule.ID)
 		}
 
 		gotRuleWithUpdates := apiRuleWithUpdates{}
@@ -173,7 +173,7 @@ func TestHandler(t *testing.T) {
 			lr := listGroupsResponse{}
 			getResp(t, ts.URL+url, &lr, 200)
 			if length := len(lr.Data.Groups); length != expGroups {
-				t.Errorf("expected %d groups got %d", expGroups, length)
+				t.Fatalf("expected %d groups got %d", expGroups, length)
 			}
 			if len(lr.Data.Groups) < 1 {
 				return
@@ -183,7 +183,7 @@ func TestHandler(t *testing.T) {
 				rulesN += len(gr.Rules)
 			}
 			if rulesN != expRules {
-				t.Errorf("expected %d rules got %d", expRules, rulesN)
+				t.Fatalf("expected %d rules got %d", expRules, rulesN)
 			}
 		}
 
@@ -241,23 +241,23 @@ func TestEmptyResponse(t *testing.T) {
 	ts := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) { rhWithNoGroups.handler(w, r) }))
 	defer ts.Close()
 
-	getResp := func(t *testing.T, url string, to interface{}, code int) {
+	getResp := func(t *testing.T, url string, to any, code int) {
 		t.Helper()
 		resp, err := http.Get(url)
 		if err != nil {
 			t.Fatalf("unexpected err %s", err)
 		}
 		if code != resp.StatusCode {
-			t.Errorf("unexpected status code %d want %d", resp.StatusCode, code)
+			t.Fatalf("unexpected status code %d want %d", resp.StatusCode, code)
 		}
 		defer func() {
 			if err := resp.Body.Close(); err != nil {
-				t.Errorf("err closing body %s", err)
+				t.Fatalf("err closing body %s", err)
 			}
 		}()
 		if to != nil {
 			if err = json.NewDecoder(resp.Body).Decode(to); err != nil {
-				t.Errorf("unexpected err %s", err)
+				t.Fatalf("unexpected err %s", err)
 			}
 		}
 	}
@@ -266,13 +266,13 @@ func TestEmptyResponse(t *testing.T) {
 		lr := listAlertsResponse{}
 		getResp(t, ts.URL+"/api/v1/alerts", &lr, 200)
 		if lr.Data.Alerts == nil {
-			t.Errorf("expected /api/v1/alerts response to have non-nil data")
+			t.Fatalf("expected /api/v1/alerts response to have non-nil data")
 		}
 
 		lr = listAlertsResponse{}
 		getResp(t, ts.URL+"/vmalert/api/v1/alerts", &lr, 200)
 		if lr.Data.Alerts == nil {
-			t.Errorf("expected /api/v1/alerts response to have non-nil data")
+			t.Fatalf("expected /api/v1/alerts response to have non-nil data")
 		}
 	})
 
@@ -280,13 +280,13 @@ func TestEmptyResponse(t *testing.T) {
 		lr := listGroupsResponse{}
 		getResp(t, ts.URL+"/api/v1/rules", &lr, 200)
 		if lr.Data.Groups == nil {
-			t.Errorf("expected /api/v1/rules response to have non-nil data")
+			t.Fatalf("expected /api/v1/rules response to have non-nil data")
 		}
 
 		lr = listGroupsResponse{}
 		getResp(t, ts.URL+"/vmalert/api/v1/rules", &lr, 200)
 		if lr.Data.Groups == nil {
-			t.Errorf("expected /api/v1/rules response to have non-nil data")
+			t.Fatalf("expected /api/v1/rules response to have non-nil data")
 		}
 	})
 

app/vmalert/web_types.go

@@ -183,7 +183,7 @@ func (ar apiRule) WebLink() string {
 		paramGroupID, ar.GroupID, paramRuleID, ar.ID)
 }
 
-func ruleToAPI(r interface{}) apiRule {
+func ruleToAPI(r any) apiRule {
 	if ar, ok := r.(*rule.AlertingRule); ok {
 		return alertingToAPI(ar)
 	}

app/vmalert/web_types_test.go

@@ -13,11 +13,11 @@ func TestUrlValuesToStrings(t *testing.T) {
 	res := urlValuesToStrings(mapQueryParams)
 
 	if len(res) != len(expectedRes) {
-		t.Errorf("Expected length %d, but got %d", len(expectedRes), len(res))
+		t.Fatalf("Expected length %d, but got %d", len(expectedRes), len(res))
 	}
 	for ind, val := range expectedRes {
 		if val != res[ind] {
-			t.Errorf("Expected %v; but got %v", val, res[ind])
+			t.Fatalf("Expected %v; but got %v", val, res[ind])
 		}
 	}
 }

app/vmauth/auth_config.go

@@ -83,6 +83,7 @@ type UserInfo struct {
 
 	concurrencyLimitCh      chan struct{}
 	concurrencyLimitReached *metrics.Counter
+	overrideHostHeader      bool
 
 	rt http.RoundTripper
 
@@ -128,7 +129,7 @@ type Header struct {
 }
 
 // UnmarshalYAML unmarshals h from f.
-func (h *Header) UnmarshalYAML(f func(interface{}) error) error {
+func (h *Header) UnmarshalYAML(f func(any) error) error {
 	var s string
 	if err := f(&s); err != nil {
 		return err
@@ -145,10 +146,19 @@ func (h *Header) UnmarshalYAML(f func(interface{}) error) error {
 }
 
 // MarshalYAML marshals h to yaml.
-func (h *Header) MarshalYAML() (interface{}, error) {
+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.
@@ -191,7 +201,7 @@ type QueryArg struct {
 }
 
 // UnmarshalYAML unmarshals qa from yaml.
-func (qa *QueryArg) UnmarshalYAML(f func(interface{}) error) error {
+func (qa *QueryArg) UnmarshalYAML(f func(any) error) error {
 	var s string
 	if err := f(&s); err != nil {
 		return err
@@ -220,7 +230,7 @@ func (qa *QueryArg) UnmarshalYAML(f func(interface{}) error) error {
 }
 
 // MarshalYAML marshals qa to yaml.
-func (qa *QueryArg) MarshalYAML() (interface{}, error) {
+func (qa *QueryArg) MarshalYAML() (any, error) {
 	return qa.sOriginal, nil
 }
 
@@ -253,7 +263,7 @@ type URLPrefix struct {
 	nextDiscoveryDeadline atomic.Uint64
 
 	// vOriginal contains the original yaml value for URLPrefix.
-	vOriginal interface{}
+	vOriginal any
 }
 
 func (up *URLPrefix) setLoadBalancingPolicy(loadBalancingPolicy string) error {
@@ -344,11 +354,12 @@ func (up *URLPrefix) discoverBackendAddrsIfNeeded() {
 			// ips for the given host have been already discovered
 			continue
 		}
+
 		var resolvedAddrs []string
 		if strings.HasPrefix(host, "srv+") {
 			// The host has the format 'srv+realhost'. Strip 'srv+' prefix before performing the lookup.
-			host = strings.TrimPrefix(host, "srv+")
-			_, addrs, err := netutil.Resolver.LookupSRV(ctx, "", "", host)
+			srvHost := strings.TrimPrefix(host, "srv+")
+			_, addrs, err := netutil.Resolver.LookupSRV(ctx, "", "", srvHost)
 			if err != nil {
 				logger.Warnf("cannot discover backend SRV records for %s: %s; use it literally", bu, err)
 				resolvedAddrs = []string{host}
@@ -380,7 +391,6 @@ func (up *URLPrefix) discoverBackendAddrsIfNeeded() {
 	var busNew []*backendURL
 	for _, bu := range up.busOriginal {
 		host := bu.Hostname()
-		host = strings.TrimPrefix(host, "srv+")
 		port := bu.Port()
 		for _, addr := range hostToAddrs[host] {
 			buCopy := *bu
@@ -487,8 +497,8 @@ func getLeastLoadedBackendURL(bus []*backendURL, atomicCounter *atomic.Uint32) *
 }
 
 // UnmarshalYAML unmarshals up from yaml.
-func (up *URLPrefix) UnmarshalYAML(f func(interface{}) error) error {
-	var v interface{}
+func (up *URLPrefix) UnmarshalYAML(f func(any) error) error {
+	var v any
 	if err := f(&v); err != nil {
 		return err
 	}
@@ -498,7 +508,7 @@ func (up *URLPrefix) UnmarshalYAML(f func(interface{}) error) error {
 	switch x := v.(type) {
 	case string:
 		urls = []string{x}
-	case []interface{}:
+	case []any:
 		if len(x) == 0 {
 			return fmt.Errorf("`url_prefix` must contain at least a single url")
 		}
@@ -528,7 +538,7 @@ func (up *URLPrefix) UnmarshalYAML(f func(interface{}) error) error {
 }
 
 // MarshalYAML marshals up to yaml.
-func (up *URLPrefix) MarshalYAML() (interface{}, error) {
+func (up *URLPrefix) MarshalYAML() (any, error) {
 	return up.vOriginal, nil
 }
 
@@ -552,7 +562,7 @@ func (r *Regex) match(s string) bool {
 }
 
 // UnmarshalYAML implements yaml.Unmarshaler
-func (r *Regex) UnmarshalYAML(f func(interface{}) error) error {
+func (r *Regex) UnmarshalYAML(f func(any) error) error {
 	var s string
 	if err := f(&s); err != nil {
 		return err
@@ -569,7 +579,7 @@ func (r *Regex) UnmarshalYAML(f func(interface{}) error) error {
 }
 
 // MarshalYAML implements yaml.Marshaler.
-func (r *Regex) MarshalYAML() (interface{}, error) {
+func (r *Regex) MarshalYAML() (any, error) {
 	return r.sOriginal, nil
 }
 
@@ -686,22 +696,15 @@ func loadAuthConfig() (bool, error) {
 	}
 	logger.Infof("loaded information about %d users from -auth.config=%q", len(m), *authConfigPath)
 
-	prevAc := authConfig.Load()
-	if prevAc != nil {
-		metrics.UnregisterSet(prevAc.ms)
+	acPrev := authConfig.Load()
+	if acPrev != nil {
+		metrics.UnregisterSet(acPrev.ms, true)
 	}
 	metrics.RegisterSet(ac.ms)
+
 	authConfig.Store(ac)
 	authConfigData.Store(&data)
 	authUsers.Store(&m)
-	if prevAc != nil {
-		// explicilty unregister metrics, since all summary type metrics
-		// are registered at global state of metrics package
-		// and must be removed from it to release memory.
-		// Metrics must be unregistered only after atomic.Value.Store calls above
-		// Otherwise it may lead to metric gaps, since UnregisterAllMetrics is slow operation
-		prevAc.ms.UnregisterAllMetrics()
-	}
 
 	return true, nil
 }
@@ -738,6 +741,7 @@ 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 {
@@ -802,6 +806,7 @@ 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 {
@@ -1012,8 +1017,6 @@ func (up *URLPrefix) sanitizeAndInitialize() error {
 		}
 	}
 	up.bus.Store(&bus)
-	up.nextDiscoveryDeadline.Store(0)
-	up.n.Store(0)
 
 	return nil
 }