fixes makefile after cherry-pick

* disable docker scan

* disable nightly, enable auto release to sandbox

* remove whitespace

(cherry picked from commit d4099a75be)

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -5,10 +5,10 @@ body:
   - type: markdown
     attributes:
       value: |
-        Before filling a bug report it would be great to [upgrade](https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#how-to-upgrade) 
-        to [the latest available release](https://github.com/VictoriaMetrics/VictoriaMetrics/releases/latest)
+        Before filling a bug report it would be great to [upgrade](https://docs.victoriametrics.com/#how-to-upgrade) 
+        to [the latest available release](https://github.com/VictoriaMetrics/VictoriaMetrics/releases)
         and verify whether the bug is reproducible there.
-        It's also recommended to read the [troubleshooting docs](https://docs.victoriametrics.com/victoriametrics/troubleshooting/) first.
+        It's also recommended to read the [troubleshooting docs](https://docs.victoriametrics.com/Troubleshooting.html) first.
   - type: textarea
     id: describe-the-bug
     attributes:
@@ -60,12 +60,12 @@ body:
   
         For VictoriaMetrics health-state issues please provide full-length screenshots
         of Grafana dashboards if possible:
-          * [Grafana dashboard for single-node VictoriaMetrics](https://grafana.com/grafana/dashboards/10229)
-          * [Grafana dashboard for VictoriaMetrics cluster](https://grafana.com/grafana/dashboards/11176)
+          * [Grafana dashboard for single-node VictoriaMetrics](https://grafana.com/grafana/dashboards/10229-victoriametrics/)
+          * [Grafana dashboard for VictoriaMetrics cluster](https://grafana.com/grafana/dashboards/11176-victoriametrics-cluster/)
         
         See how to setup monitoring here:
-          * [monitoring for single-node VictoriaMetrics](https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#monitoring)
-          * [monitoring for VictoriaMetrics cluster](https://docs.victoriametrics.com/victoriametrics/cluster-victoriametrics/#monitoring)
+          * [monitoring for single-node VictoriaMetrics](https://docs.victoriametrics.com/#monitoring)
+          * [monitoring for VictoriaMetrics cluster](https://docs.victoriametrics.com/Cluster-VictoriaMetrics.html#monitoring)
     validations:
       required: false
   - type: textarea

.github/ISSUE_TEMPLATE/question.yml

@@ -24,9 +24,9 @@ body:
       label: Troubleshooting docs
       description: I am familiar with the following troubleshooting docs
       options:
-        - label: General - https://docs.victoriametrics.com/victoriametrics/troubleshooting/
+        - label: General - https://docs.victoriametrics.com/Troubleshooting.html
           required: false
-        - label: vmagent - https://docs.victoriametrics.com/victoriametrics/vmagent/#troubleshooting
-          required: false
-        - label: vmalert - https://docs.victoriametrics.com/victoriametrics/vmalert/#troubleshooting
+        - label: vmagent - https://docs.victoriametrics.com/vmagent.html#troubleshooting
           required: false
+        - label: vmalert - https://docs.victoriametrics.com/vmalert.html#troubleshooting
+          required: false

.github/copilot-instructions.md

@@ -1,23 +0,0 @@
-# Project Overview
-
-VictoriaMetrics is a fast, cost-saving, and scalable solution for monitoring and managing time series data. It delivers high performance and reliability, making it an ideal choice for businesses of all sizes.
-
-## Folder Structure
-
-- `/app`: Contains the compilable binaries.
-- `/lib`: Contains the golang reusable libraries
-- `/docs/victoriametrics`: Contains documentation for the project.
-- `/apptest/tests`: Contains integration tests.
-
-## Libraries and Frameworks
-
-- Backend: Golang, no framework. Use third-party libraries sparingly.
-- Frontend: React.
-
-## Code review guidelines
-
-Ensure the feature or bugfix includes a changelog entry in /docs/victoriametrics/changelog/CHANGELOG.md.
-Verify the entry is under the ## tip section and matches the structure and style of existing entries.
-Chore-only changes may be omitted from the changelog.
-
-

.github/dependabot.yml

@@ -8,12 +8,10 @@ updates:
     directory: "/"
     schedule:
       interval: "weekly"
-    open-pull-requests-limit: 0
   - package-ecosystem: "bundler"
     directory: "/docs"
     schedule:
-      interval: "weekly"
-    open-pull-requests-limit: 0
+      interval: "daily"
   - package-ecosystem: "gomod"
     directory: "/app/vmui/packages/vmui/web"
     schedule:

.github/pull_request_template.md

@@ -1,10 +0,0 @@
-### Describe Your Changes
-
-Please provide a brief description of the changes you made. Be as specific as possible to help others understand the purpose and impact of your modifications.
-
-### Checklist
-
-The following checks are **mandatory**:
-
-- [ ] My change adheres to [VictoriaMetrics contributing guidelines](https://docs.victoriametrics.com/victoriametrics/contributing/#pull-request-checklist).
-- [ ] My change adheres to [VictoriaMetrics development goals](https://docs.victoriametrics.com/victoriametrics/goals/).

.github/workflows/build.yml

@@ -1,69 +0,0 @@
-name: build
-
-on:
-  push:
-    branches:
-      - cluster
-      - master
-    paths:
-      - '**.go'
-      - '**/Dockerfile'
-      - '**/Makefile'
-      - '!app/vmui/**'
-      - '.github/workflows/build.yml'
-  pull_request:
-    branches:
-      - cluster
-      - master
-    paths:
-      - '**.go'
-      - '**/Dockerfile'
-      - '**/Makefile'
-      - '!app/vmui/**'
-      - '.github/workflows/build.yml'
-
-permissions:
-  contents: read
-
-concurrency:
-  cancel-in-progress: true
-  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
-
-jobs:
-  build:
-    name: ${{ matrix.os }}-${{ matrix.arch }}
-    runs-on: ubuntu-latest
-    strategy:
-      fail-fast: false
-      matrix:
-        include:
-          - os: linux
-            arch: amd64
-          - os: linux
-            arch: arm64
-          - os: linux
-            arch: arm
-          - os: linux
-            arch: ppc64le
-          - os: linux
-            arch: 386
-          - os: freebsd
-            arch: amd64
-          - os: openbsd
-            arch: amd64
-    steps:
-      - name: Code checkout
-        uses: actions/checkout@v5
-
-      - name: Setup Go
-        id: go
-        uses: actions/setup-go@v5
-        with:
-          cache-dependency-path: |
-            go.sum
-            Makefile
-            app/**/Makefile
-          go-version: stable
-
-      - name: Build vmcluster for ${{ matrix.os }}-${{ matrix.arch }}
-        run: make vmcluster-${{ matrix.os }}-${{ matrix.arch }}

.github/workflows/check-licenses.yml

@@ -14,25 +14,13 @@ jobs:
     name: Build
     runs-on: ubuntu-latest
     steps:
+      - name: Setup Go
+        uses: actions/setup-go@main
+        with:
+          go-version: 1.20.4
+        id: go
       - name: Code checkout
         uses: actions/checkout@master
-
-      - 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-${{ runner.os }}-check-licenses-${{ steps.go.outputs.go-version }}-${{ hashFiles('go.sum', 'Makefile', 'app/**/Makefile') }}
-          restore-keys: go-artifacts-${{ runner.os }}-check-licenses-
-
       - name: Check License
-        run: make check-licenses
+        run: |
+          make check-licenses

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

@@ -1,62 +0,0 @@
-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@v5
-
-      - 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.yml

@@ -0,0 +1,46 @@
+name: "CodeQL - JS"
+
+on:
+  push:
+    branches: [master, cluster]
+    paths:
+      - "**.js"
+  pull_request:
+    # The branches below must be a subset of the branches above
+    branches: [master, cluster]
+    paths:
+      - "**.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: ["javascript"]
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v3
+
+      - name: Initialize CodeQL
+        uses: github/codeql-action/init@v2
+        with:
+          languages: ${{ matrix.language }}
+
+      - name: Perform CodeQL Analysis
+        uses: github/codeql-action/analyze@v2
+        with:
+          category: "javascript"

.github/workflows/codeql-analysis.yml

@@ -0,0 +1,92 @@
+# 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@v3
+
+      - name: Set up Go
+        uses: actions/setup-go@v4
+        with:
+          go-version: 1.20.4
+          check-latest: true
+          cache: true
+        if: ${{ matrix.language == 'go' }}
+
+      # Initializes the CodeQL tools for scanning.
+      - name: Initialize CodeQL
+        uses: github/codeql-action/init@v2
+        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@v2
+
+      # 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@v2

.github/workflows/docs.yaml

@@ -1,57 +0,0 @@
-name: publish-docs
-on:
-  push:
-    branches:
-      - 'master'
-    paths:
-      - 'docs/**'
-      - '.github/workflows/docs.yaml'
-  workflow_dispatch: {}
-permissions:
-  contents: read  # This is required for actions/checkout and to commit back image update
-  deployments: write
-jobs:
-  build:
-    name: Build
-    runs-on: ubuntu-latest
-    steps:
-      - name: Code checkout
-        uses: actions/checkout@v5
-        with:
-          path: __vm
-
-      - name: Checkout private code
-        uses: actions/checkout@v5
-        with:
-          repository: VictoriaMetrics/vmdocs
-          token: ${{ secrets.VM_BOT_GH_TOKEN }}
-          path: __vm-docs
-
-      - name: Import GPG key
-        uses: crazy-max/ghaction-import-gpg@v6
-        id: import-gpg
-        with:
-          gpg_private_key: ${{ secrets.VM_BOT_GPG_PRIVATE_KEY }}
-          passphrase: ${{ secrets.VM_BOT_PASSPHRASE }}
-          git_user_signingkey: true
-          git_commit_gpgsign: true
-          git_config_global: true
-
-      - name: Copy docs
-        id: update
-        run: |
-          find docs -type d -maxdepth 1 -mindepth 1 -exec \
-            sh -c 'rsync -zarvh --delete {}/ ../__vm-docs/content/$(basename {})/' \;
-          echo "SHORT_SHA=$(git rev-parse --short $GITHUB_SHA)" >> $GITHUB_OUTPUT
-        working-directory: __vm
-
-      - name: Push to vmdocs
-        run: |
-          git config --global user.name "${{ steps.import-gpg.outputs.email }}"
-          git config --global user.email "${{ steps.import-gpg.outputs.email }}"
-          if [[ -n $(git status --porcelain) ]]; then
-            git add .
-            git commit -S -m "sync docs with VictoriaMetrics/VictoriaMetrics commit: ${{ steps.update.outputs.SHORT_SHA }}"
-            git push
-          fi
-        working-directory: __vm-docs

.github/workflows/main.yml

@@ -0,0 +1,95 @@
+name: main
+on:
+  push:
+    branches:
+      - master
+      - cluster
+    paths-ignore:
+      - "docs/**"
+      - "**.md"
+  pull_request:
+    branches:
+      - master
+      - cluster
+    paths-ignore:
+      - "docs/**"
+      - "**.md"
+permissions:
+  contents: read
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  lint:
+    name: lint
+    runs-on: ubuntu-latest
+    steps:
+      - name: Code checkout
+        uses: actions/checkout@v3
+
+      - name: Setup Go
+        uses: actions/setup-go@v4
+        with:
+          go-version: 1.20.4
+          check-latest: true
+          cache: true
+
+      - name: Dependencies
+        run: |
+          make install-golangci-lint
+          make check-all
+          git diff --exit-code
+
+  test:
+    needs: lint
+    strategy:
+      matrix:
+        scenario: ["test-full", "test-pure", "test-full-386"]
+    name: test
+    runs-on: ubuntu-latest
+    steps:
+      - name: Code checkout
+        uses: actions/checkout@v3
+
+      - name: Setup Go
+        uses: actions/setup-go@v4
+        with:
+          go-version: 1.20.4
+          check-latest: true
+          cache: true
+
+      - name: run tests
+        run: |
+          make ${{ matrix.scenario}}
+
+      - name: Publish coverage
+        uses: codecov/codecov-action@v3
+        with:
+          file: ./coverage.txt
+
+  build:
+    needs: test
+    name: build
+    runs-on: ubuntu-latest
+    steps:
+      - name: Code checkout
+        uses: actions/checkout@v3
+
+      - name: Setup Go
+        id: go
+        uses: actions/setup-go@v4
+        with:
+          go-version: 1.20.4
+          check-latest: true
+          cache: true
+
+      - uses: actions/cache@v3
+        with:
+          path: gocache-for-docker
+          key:  gocache-docker-${{ runner.os }}-${{ steps.go.outputs.go-version }}-${{ hashFiles('go.mod') }}
+
+      - name: Build
+        run: |
+          make vmcluster-crossbuild

.github/workflows/test.yml

@@ -1,113 +0,0 @@
-name: test
-
-on:
-  push:
-    branches:
-      - cluster
-      - master
-    paths:
-      - '**.go'
-      - 'go.*'
-      - '.github/workflows/main.yml'
-  pull_request:
-    branches:
-      - cluster
-      - master
-    paths:
-      - '**.go'
-      - 'go.*'
-      - '.github/workflows/main.yml'
-
-permissions:
-  contents: read
-
-concurrency:
-  cancel-in-progress: true
-  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
-
-
-jobs:
-  lint:
-    name: lint
-    runs-on: ubuntu-latest
-    steps:
-      - name: Code checkout
-        uses: actions/checkout@v5
-
-      - name: Setup Go
-        id: go
-        uses: actions/setup-go@v5
-        with:
-          cache-dependency-path: |
-            go.sum
-            Makefile
-            app/**/Makefile
-          go-version: stable
-
-
-      - name: Cache golangci-lint
-        uses: actions/cache@v4
-        with:
-          path: |
-            ~/.cache/golangci-lint
-            ~/go/bin
-          key: golangci-lint-${{ runner.os }}-${{ hashFiles('.golangci.yml') }}
-
-      - name: Run check-all
-        run: |
-          make check-all
-          git diff --exit-code
-
-  unit:
-    name: unit
-    runs-on: ubuntu-latest
-
-    strategy:
-      matrix:
-        scenario:
-          - 'test-full'
-          - 'test-full-386'
-          - 'test-pure'
-
-    steps:
-      - name: Code checkout
-        uses: actions/checkout@v5
-
-      - name: Setup Go
-        id: go
-        uses: actions/setup-go@v5
-        with:
-          cache-dependency-path: |
-            go.sum
-            Makefile
-            app/**/Makefile
-          go-version: stable
-
-      - name: Run tests
-        run: GOGC=10 make ${{ matrix.scenario}}
-
-      - name: Publish coverage
-        uses: codecov/codecov-action@v5
-        with:
-          files: ./coverage.txt
-
-  integration:
-    name: integration
-    runs-on: ubuntu-latest
-
-    steps:
-      - name: Code checkout
-        uses: actions/checkout@v5
-
-      - name: Setup Go
-        id: go
-        uses: actions/setup-go@v5
-        with:
-          cache-dependency-path: |
-            go.sum
-            Makefile
-            app/**/Makefile
-          go-version: stable
-
-      - name: Run integration tests
-        run: make integration-test

.github/workflows/update-sandbox.yml

@@ -0,0 +1,77 @@
+name: sandbox-release
+on:
+  release:
+    types: [published]
+permissions:
+  contents: write
+jobs:
+  deploy-sandbox:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Check out code
+        uses: actions/checkout@v3
+        with:
+          repository: VictoriaMetrics/ops
+          ref: master
+          token: ${{ secrets.VM_BOT_GH_TOKEN }}
+
+      - name: Import GPG key
+        id: import-gpg
+        uses: crazy-max/ghaction-import-gpg@v5
+        with:
+          gpg_private_key: ${{ secrets.VM_BOT_GPG_PRIVATE_KEY }}
+          passphrase: ${{ secrets.VM_BOT_PASSPHRASE }}
+          git_user_signingkey: true
+          git_commit_gpgsign: true
+
+      - name: update image tag
+        uses: fjogeleit/yaml-update-action@main
+        with:
+          valueFile: 'gcp-test/sandbox/manifests/benchmark-vm/vmcluster.yaml'
+          commitChange: false
+          createPR: false
+          changes: |
+            {
+              "gcp-test/sandbox/manifests/benchmark-vm/vmcluster.yaml": {
+                "spec.vminsert.image.tag": "${{ github.event.release.tag_name }}-enterprise-cluster",
+                "spec.vmselect.image.tag": "${{ github.event.release.tag_name }}-enterprise-cluster",
+                "spec.vmstorage.image.tag": "${{ github.event.release.tag_name }}-enterprise-cluster"
+              },
+              "gcp-test/sandbox/manifests/benchmark-vm/vmsingle.yaml": {
+                "spec.image.tag": "${{ github.event.release.tag_name }}-enterprise"
+              },
+              "gcp-test/sandbox/manifests/monitoring/monitoring-vmagent.yaml": {
+                "spec.image.tag": "${{ github.event.release.tag_name }}"
+              },
+              "gcp-test/sandbox/manifests/monitoring/monitoring-vmcluster.yaml": {
+                "spec.vminsert.image.tag": "${{ github.event.release.tag_name }}-enterprise-cluster",
+                "spec.vmselect.image.tag": "${{ github.event.release.tag_name }}-enterprise-cluster",
+                "spec.vmstorage.image.tag": "${{ github.event.release.tag_name }}-enterprise-cluster"
+              },
+              "gcp-test/sandbox/manifests/monitoring/vmalert.yaml": {
+                "spec.image.tag": "${{ github.event.release.tag_name }}-enterprise"
+              }
+            }
+
+      - name: commit changes
+        run: |
+          git config --global user.name "${{ steps.import-gpg.outputs.email }}"
+          git config --global user.email "${{ steps.import-gpg.outputs.email }}"
+          git add .
+          git commit -S -m "Deploy image tag ${RELEASE_TAG} to sandbox"
+        env:
+          RELEASE_TAG: ${{ github.event.release.tag_name }}
+
+      - name: Create Pull Request
+        uses: peter-evans/create-pull-request@v5
+        with:
+          author: ${{ github.actor }} <${{ github.actor }}@users.noreply.github.com>
+          branch: release-automation
+          token: ${{ secrets.VM_BOT_GH_TOKEN }}
+          delete-branch: true
+          title: "release ${{ github.event.release.tag_name }}"
+          body: |
+            Release [${{ github.event.release.tag_name }}](https://github.com/VictoriaMetrics/VictoriaMetrics/releases/tag/${{ github.event.release.tag_name }}) to sandbox
+        
+            > Auto-generated by `Github Actions Bot`
+    

.github/workflows/vmui.yml

@@ -1,82 +0,0 @@
-name: vmui
-
-on:
-  push:
-    branches:
-      - cluster
-      - master
-    paths:
-      - 'app/vmui/packages/vmui/**'
-      - '.github/workflows/vmui.yml'
-  pull_request:
-    branches:
-      - cluster
-      - master
-    paths:
-      - 'app/vmui/packages/vmui/**'
-      - '.github/workflows/vmui.yml'
-
-permissions:
-  contents: read
-  packages: read
-  pull-requests: read
-  checks: write
-
-concurrency:
-  cancel-in-progress: true
-  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
-
-jobs:
-  vmui-checks:
-    name: VMUI Checks (lint, test, typecheck)
-    runs-on: ubuntu-latest
-    steps:
-      - name: Code checkout
-        uses: actions/checkout@v5
-
-      - name: Setup Node
-        uses: actions/setup-node@v4
-        with:
-          node-version: '24.x'
-
-      - name: Cache node-modules
-        uses: actions/cache@v4
-        with:
-          path: |
-            app/vmui/packages/vmui/node_modules
-          key: vmui-artifacts-${{ runner.os }}-${{ hashFiles('package-lock.json') }}
-          restore-keys: vmui-artifacts-${{ runner.os }}-
-
-      - name: Run lint
-        id: lint
-        run: make vmui-lint
-        continue-on-error: true
-
-      - name: Run tests
-        id: test
-        run: make vmui-test
-        continue-on-error: true
-
-      - name: Run typecheck
-        id: typecheck
-        run: make vmui-typecheck
-        continue-on-error: true
-
-      - name: Annotate Code Linting Results
-        uses: ataylorme/eslint-annotate-action@v3
-        with:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-          report-json: app/vmui/packages/vmui/vmui-lint-report.json
-
-      - name: Check overall status
-        run: |
-          echo "Lint status: ${{ steps.lint.outcome }}"
-          echo "Test status: ${{ steps.test.outcome }}"
-          echo "Typecheck status: ${{ steps.typecheck.outcome }}"
-
-          if [[ "${{ steps.lint.outcome }}" == "failure" || "${{ steps.test.outcome }}" == "failure" || "${{ steps.typecheck.outcome }}" == "failure" ]]; then
-            echo "One or more checks failed"
-            exit 1
-          else
-            echo "All checks passed"
-          fi

.gitignore

@@ -7,12 +7,9 @@
 .vscode
 *.test
 *.swp
-/vmdocs
 /gocache-for-docker
-/victoria-logs-data
 /victoria-metrics-data
 /vmagent-remotewrite-data
-/vlagent-remotewritewrite
 /vmstorage-data
 /vmselect-cache
 /package/temp-deb-*
@@ -23,9 +20,4 @@
 Gemfile.lock
 /_site
 _site
-*.tmp
-/docs/.jekyll-metadata
-coverage.txt
-cspell.json
-*~
-deployment/docker/provisioning/plugins/
+*.tmp

.golangci.yml

@@ -1,29 +1,19 @@
-version: "2"
+run:
+  timeout: 2m
+
 linters:
-  settings:
-    errcheck:
-      exclude-functions:
-        - fmt.Fprintf
-        - fmt.Fprint
-        - (net/http.ResponseWriter).Write
-  exclusions:
-    generated: lax
-    presets:
-      - common-false-positives
-      - legacy
-      - std-error-handling
-    rules:
-      - linters:
-          - staticcheck
-        text: 'SA(4003|1019|5011):'
-    paths:
-      - third_party$
-      - builtin$
-      - examples$
-formatters:
-  exclusions:
-    generated: lax
-    paths:
-      - third_party$
-      - builtin$
-      - examples$
+  enable:
+  - revive
+
+issues:
+  exclude-rules:
+  - linters:
+    - staticcheck
+    text: "SA(4003|1019|5011):"
+  include:
+  - EXC0012
+  - EXC0014
+
+linters-settings:
+  errcheck:
+    exclude: ./errcheck_excludes.txt

.wwhrd.yml

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

CODE_OF_CONDUCT_RU.md

@@ -0,0 +1,120 @@
+
+# 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>.

CONTRIBUTING.md

@@ -1 +1,16 @@
-The document has been moved [here](https://docs.victoriametrics.com/victoriametrics/contributing/).
+If you like VictoriaMetrics and want to contribute, then we need the following:
+
+- Filing issues and feature requests [here](https://github.com/VictoriaMetrics/VictoriaMetrics/issues).
+- Spreading a word about VictoriaMetrics: conference talks, articles, comments, experience sharing with colleagues.
+- Updating documentation.
+
+We are open to third-party pull requests provided they follow [KISS design principle](https://en.wikipedia.org/wiki/KISS_principle):
+
+- Prefer simple code and architecture.
+- Avoid complex abstractions.
+- Avoid magic code and fancy algorithms.
+- Avoid [big external dependencies](https://medium.com/@valyala/stripping-dependency-bloat-in-victoriametrics-docker-image-983fb5912b0d).
+- Minimize the number of moving parts in the distributed system.
+- Avoid automated decisions, which may hurt cluster availability, consistency or performance.
+
+Adhering `KISS` principle simplifies the resulting code and architecture, so it can be reviewed, understood and verified by many people.

LICENSE

@@ -175,7 +175,7 @@
 
    END OF TERMS AND CONDITIONS
 
-   Copyright 2019-2025 VictoriaMetrics, Inc.
+   Copyright 2019-2023 VictoriaMetrics, Inc.
 
    Licensed under the Apache License, Version 2.0 (the "License");
    you may not use this file except in compliance with the License.

Makefile

@@ -1,33 +1,22 @@
 PKG_PREFIX := github.com/VictoriaMetrics/VictoriaMetrics
 
-MAKE_CONCURRENCY ?= $(shell getconf _NPROCESSORS_ONLN)
-MAKE_PARALLEL := $(MAKE) -j $(MAKE_CONCURRENCY)
 DATEINFO_TAG ?= $(shell date -u +'%Y%m%d-%H%M%S')
 BUILDINFO_TAG ?= $(shell echo $$(git describe --long --all | tr '/' '-')$$( \
 	      git diff-index --quiet HEAD -- || echo '-dirty-'$$(git diff-index -u HEAD | openssl sha1 | cut -d' ' -f2 | cut -c 1-8)))
+LATEST_TAG ?= enterprise-cluster-latest
 
 PKG_TAG ?= $(shell git tag -l --points-at HEAD)
 ifeq ($(PKG_TAG),)
 PKG_TAG := $(BUILDINFO_TAG)
 endif
 
-EXTRA_DOCKER_TAG_SUFFIX ?=
-EXTRA_GO_BUILD_TAGS ?=
-
 GO_BUILDINFO = -X '$(PKG_PREFIX)/lib/buildinfo.Version=$(APP_NAME)-$(DATEINFO_TAG)-$(BUILDINFO_TAG)'
-TAR_OWNERSHIP ?= --owner=1000 --group=1000
-
-GOLANGCI_LINT_VERSION := 2.4.0
 
 .PHONY: $(MAKECMDGOALS)
 
 include app/*/Makefile
-include codespell/Makefile
-include docs/Makefile
 include deployment/*/Makefile
-include dashboards/Makefile
 include package/release/Makefile
-include benchmarks/Makefile
 
 all: \
 	vminsert \
@@ -82,31 +71,16 @@ vmcluster-windows-amd64: \
 	vmselect-windows-amd64 \
 	vmstorage-windows-amd64
 
-vmcluster-darwin-amd64: \
-	vminsert-darwin-amd64 \
-	vmselect-darwin-amd64 \
-	vmstorage-darwin-amd64
+vmcluster-crossbuild: \
+	vmcluster-linux-amd64 \
+	vmcluster-linux-arm64 \
+	vmcluster-linux-arm \
+	vmcluster-linux-ppc64le \
+	vmcluster-linux-386 \
+	vmcluster-freebsd-amd64 \
+	vmcluster-openbsd-amd64
 
-vmcluster-darwin-arm64: \
-	vminsert-darwin-arm64 \
-	vmselect-darwin-arm64 \
-	vmstorage-darwin-arm64
-
-# When adding a new crossbuild target, please also add it to the .github/workflows/build.yml
-crossbuild: vmcluster-crossbuild
-
-# When adding a new crossbuild target, please also add it to the .github/workflows/build.yml
-vmcluster-crossbuild:
-	$(MAKE_PARALLEL) vmcluster-linux-amd64 \
-		vmcluster-linux-arm64 \
-		vmcluster-linux-arm \
-		vmcluster-linux-ppc64le \
-		vmcluster-linux-386 \
-		vmcluster-freebsd-amd64 \
-		vmcluster-openbsd-amd64
-
-publish: \
-	publish-vminsert \
+publish: publish-vminsert \
 	publish-vmselect \
 	publish-vmstorage
 
@@ -115,64 +89,21 @@ package: \
 	package-vmselect \
 	package-vmstorage
 
-publish-final-images:
-	PKG_TAG=$(TAG) APP_NAME=victoria-metrics $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG) APP_NAME=vmagent $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG) APP_NAME=vmalert $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG) APP_NAME=vmalert-tool $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG) APP_NAME=vmauth $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG) APP_NAME=vmbackup $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG) APP_NAME=vmrestore $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG) APP_NAME=vmctl $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-cluster APP_NAME=vminsert $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-cluster APP_NAME=vmselect $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-cluster APP_NAME=vmstorage $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-enterprise APP_NAME=victoria-metrics $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-enterprise APP_NAME=vmagent $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-enterprise APP_NAME=vmalert $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-enterprise APP_NAME=vmauth $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-enterprise APP_NAME=vmbackup $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-enterprise APP_NAME=vmrestore $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-enterprise-cluster APP_NAME=vminsert $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-enterprise-cluster APP_NAME=vmselect $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-enterprise-cluster APP_NAME=vmstorage $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-enterprise APP_NAME=vmgateway $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG)-enterprise APP_NAME=vmbackupmanager $(MAKE) publish-via-docker-from-rc && \
-	PKG_TAG=$(TAG) $(MAKE) publish-latest
-
-publish-latest:
-	PKG_TAG=$(TAG) APP_NAME=victoria-metrics $(MAKE) publish-via-docker-latest && \
-	PKG_TAG=$(TAG) APP_NAME=vmagent $(MAKE) publish-via-docker-latest && \
-	PKG_TAG=$(TAG) APP_NAME=vmalert $(MAKE) publish-via-docker-latest && \
-	PKG_TAG=$(TAG) APP_NAME=vmalert-tool $(MAKE) publish-via-docker-latest && \
-	PKG_TAG=$(TAG) APP_NAME=vmauth $(MAKE) publish-via-docker-latest && \
-	PKG_TAG=$(TAG) APP_NAME=vmbackup $(MAKE) publish-via-docker-latest && \
-	PKG_TAG=$(TAG) APP_NAME=vmrestore $(MAKE) publish-via-docker-latest && \
-	PKG_TAG=$(TAG) APP_NAME=vmctl $(MAKE) publish-via-docker-latest && \
-	PKG_TAG=$(TAG)-cluster APP_NAME=vminsert $(MAKE) publish-via-docker-latest && \
-	PKG_TAG=$(TAG)-cluster APP_NAME=vmselect $(MAKE) publish-via-docker-latest && \
-	PKG_TAG=$(TAG)-cluster APP_NAME=vmstorage $(MAKE) publish-via-docker-latest && \
-	PKG_TAG=$(TAG)-enterprise APP_NAME=vmgateway $(MAKE) publish-via-docker-latest
-	PKG_TAG=$(TAG)-enterprise APP_NAME=vmbackupmanager $(MAKE) publish-via-docker-latest
-
 publish-release:
-	rm -rf bin/*
-	git checkout $(TAG) && $(MAKE) release && $(MAKE) publish && \
-		git checkout $(TAG)-cluster && $(MAKE) release && $(MAKE) publish && \
-		git checkout $(TAG)-enterprise && $(MAKE) release && $(MAKE) publish && \
-		git checkout $(TAG)-enterprise-cluster && $(MAKE) release && $(MAKE) publish
+	git checkout $(TAG) && LATEST_TAG=stable $(MAKE) release publish && \
+		git checkout $(TAG)-cluster && LATEST_TAG=cluster-stable $(MAKE) release publish && \
+		git checkout $(TAG)-enterprise && LATEST_TAG=enterprise-stable $(MAKE) release publish && \
+		git checkout $(TAG)-enterprise-cluster && LATEST_TAG=enterprise-cluster-stable $(MAKE) release publish
 
-release:
-	$(MAKE_PARALLEL) release-vmcluster
+release: \
+	release-vmcluster
 
 release-vmcluster: \
 	release-vmcluster-linux-amd64 \
 	release-vmcluster-linux-arm64 \
 	release-vmcluster-freebsd-amd64 \
 	release-vmcluster-openbsd-amd64 \
-	release-vmcluster-windows-amd64 \
-	release-vmcluster-darwin-amd64 \
-	release-vmcluster-darwin-arm64
+	release-vmcluster-windows-amd64
 
 release-vmcluster-linux-amd64:
 	GOOS=linux GOARCH=amd64 $(MAKE) release-vmcluster-goos-goarch
@@ -189,18 +120,12 @@ release-vmcluster-openbsd-amd64:
 release-vmcluster-windows-amd64:
 	GOARCH=amd64 $(MAKE) release-vmcluster-windows-goarch
 
-release-vmcluster-darwin-amd64:
-	GOOS=darwin GOARCH=amd64 $(MAKE) release-vmcluster-goos-goarch
-
-release-vmcluster-darwin-arm64:
-	 GOOS=darwin GOARCH=arm64 $(MAKE) release-vmcluster-goos-goarch
-
 release-vmcluster-goos-goarch: \
 	vminsert-$(GOOS)-$(GOARCH)-prod \
 	vmselect-$(GOOS)-$(GOARCH)-prod \
 	vmstorage-$(GOOS)-$(GOARCH)-prod
 	cd bin && \
-		tar $(TAR_OWNERSHIP) --transform="flags=r;s|-$(GOOS)-$(GOARCH)||" -czf victoria-metrics-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
+		tar --transform="flags=r;s|-$(GOOS)-$(GOARCH)||" -czf victoria-metrics-$(GOOS)-$(GOARCH)-$(PKG_TAG).tar.gz \
 			vminsert-$(GOOS)-$(GOARCH)-prod \
 			vmselect-$(GOOS)-$(GOARCH)-prod \
 			vmstorage-$(GOOS)-$(GOARCH)-prod \
@@ -239,64 +164,53 @@ pprof-cpu:
 fmt:
 	gofmt -l -w -s ./lib
 	gofmt -l -w -s ./app
-	gofmt -l -w -s ./apptest
 
 vet:
-	GOEXPERIMENT=synctest go vet ./lib/...
+	go vet ./lib/...
 	go vet ./app/...
-	go vet ./apptest/...
 
 check-all: fmt vet golangci-lint govulncheck
 
-clean-checkers: remove-golangci-lint remove-govulncheck
-
 test:
-	GOEXPERIMENT=synctest go test ./lib/... ./app/...
+	go test ./lib/... ./app/...
 
 test-race:
-	GOEXPERIMENT=synctest go test -race ./lib/... ./app/...
+	go test -race ./lib/... ./app/...
 
 test-pure:
-	GOEXPERIMENT=synctest CGO_ENABLED=0 go test ./lib/... ./app/...
+	CGO_ENABLED=0 go test ./lib/... ./app/...
 
 test-full:
-	GOEXPERIMENT=synctest go test -coverprofile=coverage.txt -covermode=atomic ./lib/... ./app/...
+	go test -coverprofile=coverage.txt -covermode=atomic ./lib/... ./app/...
 
 test-full-386:
-	GOEXPERIMENT=synctest GOARCH=386 go test -coverprofile=coverage.txt -covermode=atomic ./lib/... ./app/...
-
-integration-test:
-	$(MAKE) apptest
-
-apptest:
-	$(MAKE) all vmctl vmbackup vmrestore
-	go test ./apptest/... -skip="^TestSingle.*"
+	GOARCH=386 go test -coverprofile=coverage.txt -covermode=atomic ./lib/... ./app/...
 
 benchmark:
-	GOEXPERIMENT=synctest go test -bench=. ./lib/...
+	go test -bench=. ./lib/...
 	go test -bench=. ./app/...
 
 benchmark-pure:
-	GOEXPERIMENT=synctest CGO_ENABLED=0 go test -bench=. ./lib/...
+	CGO_ENABLED=0 go test -bench=. ./lib/...
 	CGO_ENABLED=0 go test -bench=. ./app/...
 
 vendor-update:
-	go get -u ./lib/...
-	go get -u ./app/...
-	go mod tidy -compat=1.24
+	go get -u -d ./lib/...
+	go get -u -d ./app/...
+	go mod tidy -compat=1.19
 	go mod vendor
 
 app-local:
-	CGO_ENABLED=1 go build $(RACE) -ldflags "$(GO_BUILDINFO)" -tags "$(EXTRA_GO_BUILD_TAGS)" -o bin/$(APP_NAME)$(RACE) $(PKG_PREFIX)/app/$(APP_NAME)
+	CGO_ENABLED=1 go build $(RACE) -ldflags "$(GO_BUILDINFO)" -o bin/$(APP_NAME)$(RACE) $(PKG_PREFIX)/app/$(APP_NAME)
 
 app-local-pure:
-	CGO_ENABLED=0 go build $(RACE) -ldflags "$(GO_BUILDINFO)" -tags "$(EXTRA_GO_BUILD_TAGS)" -o bin/$(APP_NAME)-pure$(RACE) $(PKG_PREFIX)/app/$(APP_NAME)
+	CGO_ENABLED=0 go build $(RACE) -ldflags "$(GO_BUILDINFO)" -o bin/$(APP_NAME)-pure$(RACE) $(PKG_PREFIX)/app/$(APP_NAME)
 
 app-local-goos-goarch:
-	CGO_ENABLED=$(CGO_ENABLED) GOOS=$(GOOS) GOARCH=$(GOARCH) go build $(RACE) -ldflags "$(GO_BUILDINFO)" -tags "$(EXTRA_GO_BUILD_TAGS)" -o bin/$(APP_NAME)-$(GOOS)-$(GOARCH)$(RACE) $(PKG_PREFIX)/app/$(APP_NAME)
+	CGO_ENABLED=$(CGO_ENABLED) GOOS=$(GOOS) GOARCH=$(GOARCH) go build $(RACE) -ldflags "$(GO_BUILDINFO)" -o bin/$(APP_NAME)-$(GOOS)-$(GOARCH)$(RACE) $(PKG_PREFIX)/app/$(APP_NAME)
 
 app-local-windows-goarch:
-	CGO_ENABLED=0 GOOS=windows GOARCH=$(GOARCH) go build $(RACE) -ldflags "$(GO_BUILDINFO)" -tags "$(EXTRA_GO_BUILD_TAGS)" -o bin/$(APP_NAME)-windows-$(GOARCH)$(RACE).exe $(PKG_PREFIX)/app/$(APP_NAME)
+	CGO_ENABLED=0 GOOS=windows GOARCH=$(GOARCH) go build $(RACE) -ldflags "$(GO_BUILDINFO)" -o bin/$(APP_NAME)-windows-$(GOARCH)$(RACE).exe $(PKG_PREFIX)/app/$(APP_NAME)
 
 quicktemplate-gen: install-qtc
 	qtc
@@ -306,13 +220,10 @@ install-qtc:
 
 
 golangci-lint: install-golangci-lint
-	GOEXPERIMENT=synctest golangci-lint run
+	golangci-lint run
 
 install-golangci-lint:
-	which golangci-lint && (golangci-lint --version | grep -q $(GOLANGCI_LINT_VERSION)) || curl -sSfL https://raw.githubusercontent.com/golangci/golangci-lint/master/install.sh | sh -s -- -b $(shell go env GOPATH)/bin v$(GOLANGCI_LINT_VERSION)
-
-remove-golangci-lint:
-	rm -rf `which 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 v1.51.2
 
 govulncheck: install-govulncheck
 	govulncheck ./...
@@ -320,11 +231,33 @@ govulncheck: install-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
+
+copy-docs:
+	echo '' > ${DST}
+	@if [ ${ORDER} -ne 0 ]; then \
+		echo "---\nsort: ${ORDER}\n---\n" > ${DST}; \
+	fi
+	cat ${SRC} >> ${DST}
+	sed -i='.tmp' 's/<img src=\"docs\//<img src=\"/' ${DST}
+	rm -rf docs/*.tmp
+
+# Copies docs for all components and adds the order tag.
+# For ORDER=0 it adds no order tag.
+# Images starting with <img src="docs/ are replaced with <img src="
+# Cluster docs are supposed to be ordered as 9th.
+# The rest of docs is ordered manually.
+docs-sync:
+	SRC=README.md DST=docs/Cluster-VictoriaMetrics.md ORDER=2 $(MAKE) copy-docs
+	SRC=app/vmagent/README.md DST=docs/vmagent.md ORDER=3 $(MAKE) copy-docs
+	SRC=app/vmalert/README.md DST=docs/vmalert.md ORDER=4 $(MAKE) copy-docs
+	SRC=app/vmauth/README.md DST=docs/vmauth.md ORDER=5 $(MAKE) copy-docs
+	SRC=app/vmbackup/README.md DST=docs/vmbackup.md ORDER=6 $(MAKE) copy-docs
+	SRC=app/vmrestore/README.md DST=docs/vmrestore.md ORDER=7 $(MAKE) copy-docs
+	SRC=app/vmctl/README.md DST=docs/vmctl.md ORDER=8 $(MAKE) copy-docs
+	SRC=app/vmgateway/README.md DST=docs/vmgateway.md ORDER=9 $(MAKE) copy-docs
+	SRC=app/vmbackupmanager/README.md DST=docs/vmbackupmanager.md ORDER=10 $(MAKE) copy-docs

SECURITY.md

@@ -2,17 +2,13 @@
 
 ## Supported Versions
 
-The following versions of VictoriaMetrics receive regular security fixes:
-
 | Version | Supported          |
 |---------|--------------------|
-| [latest release](https://docs.victoriametrics.com/victoriametrics/changelog/) | :white_check_mark: |
-| v1.102.x [LTS line](https://docs.victoriametrics.com/victoriametrics/lts-releases/) | :white_check_mark: |
-| v1.110.x [LTS line](https://docs.victoriametrics.com/victoriametrics/lts-releases/) | :white_check_mark: |
+| [latest release](https://docs.victoriametrics.com/CHANGELOG.html) | :white_check_mark: |
+| v1.87.x LTS release | :white_check_mark: |
+| v1.79.x LTS release | :white_check_mark: |
 | other releases  | :x:                |
 
-See [this page](https://victoriametrics.com/security/) for more details.
-
 ## Reporting a Vulnerability
 
-Please report any security issues to <security@victoriametrics.com>
+Please report any security issues to security@victoriametrics.com

app/victoria-logs/README.md

@@ -1 +0,0 @@
-VictoriaLogs source code has been moved to [github.com/VictoriaMetrics/VictoriaLogs](https://github.com/VictoriaMetrics/VictoriaLogs/).

app/vlagent/README.md

@@ -1 +0,0 @@
-VictoriaLogs source code has been moved to [github.com/VictoriaMetrics/VictoriaLogs](https://github.com/VictoriaMetrics/VictoriaLogs/).

app/vlinsert/README.md

@@ -1 +0,0 @@
-VictoriaLogs source code has been moved to [github.com/VictoriaMetrics/VictoriaLogs](https://github.com/VictoriaMetrics/VictoriaLogs/).

app/vlogscli/README.md

@@ -1 +0,0 @@
-VictoriaLogs source code has been moved to [github.com/VictoriaMetrics/VictoriaLogs](https://github.com/VictoriaMetrics/VictoriaLogs/).

app/vlogsgenerator/README.md

@@ -1 +0,0 @@
-VictoriaLogs source code has been moved to [github.com/VictoriaMetrics/VictoriaLogs](https://github.com/VictoriaMetrics/VictoriaLogs/).

app/vlselect/README.md

@@ -1 +0,0 @@
-VictoriaLogs source code has been moved to [github.com/VictoriaMetrics/VictoriaLogs](https://github.com/VictoriaMetrics/VictoriaLogs/).

app/vlstorage/README.md

@@ -1 +0,0 @@
-VictoriaLogs source code has been moved to [github.com/VictoriaMetrics/VictoriaLogs](https://github.com/VictoriaMetrics/VictoriaLogs/).

app/vmagent/Makefile

@@ -88,9 +88,6 @@ vmagent-linux-ppc64le:
 vmagent-linux-s390x:
 	APP_NAME=vmagent CGO_ENABLED=0 GOOS=linux GOARCH=s390x $(MAKE) app-local-goos-goarch
 
-vmagent-linux-loong64:
-	APP_NAME=vmagent CGO_ENABLED=0 GOOS=linux GOARCH=loong64 $(MAKE) app-local-goos-goarch
-
 vmagent-linux-386:
 	APP_NAME=vmagent CGO_ENABLED=0 GOOS=linux GOARCH=386 $(MAKE) app-local-goos-goarch
 

app/vmagent/common/push_ctx.go

@@ -3,27 +3,31 @@ package common
 import (
 	"sync"
 
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/cgroup"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promrelabel"
 )
 
 // PushCtx is a context used for populating WriteRequest.
 type PushCtx struct {
-	// WriteRequest contains the WriteRequest, which must be pushed later to remote storage.
-	//
-	// The actual labels and samples for the time series are stored in Labels and Samples fields.
-	WriteRequest prompb.WriteRequest
+	WriteRequest prompbmarshal.WriteRequest
 
 	// Labels contains flat list of all the labels used in WriteRequest.
-	Labels []prompb.Label
+	Labels []prompbmarshal.Label
 
 	// Samples contains flat list of all the samples used in WriteRequest.
-	Samples []prompb.Sample
+	Samples []prompbmarshal.Sample
 }
 
 // Reset resets ctx.
 func (ctx *PushCtx) Reset() {
-	ctx.WriteRequest.Reset()
+	tss := ctx.WriteRequest.Timeseries
+	for i := range tss {
+		ts := &tss[i]
+		ts.Labels = nil
+		ts.Samples = nil
+	}
+	ctx.WriteRequest.Timeseries = ctx.WriteRequest.Timeseries[:0]
 
 	promrelabel.CleanLabels(ctx.Labels)
 	ctx.Labels = ctx.Labels[:0]
@@ -35,10 +39,15 @@ func (ctx *PushCtx) Reset() {
 //
 // Call PutPushCtx when the ctx is no longer needed.
 func GetPushCtx() *PushCtx {
-	if v := pushCtxPool.Get(); v != nil {
-		return v.(*PushCtx)
+	select {
+	case ctx := <-pushCtxPoolCh:
+		return ctx
+	default:
+		if v := pushCtxPool.Get(); v != nil {
+			return v.(*PushCtx)
+		}
+		return &PushCtx{}
 	}
-	return &PushCtx{}
 }
 
 // PutPushCtx returns ctx to the pool.
@@ -46,7 +55,12 @@ func GetPushCtx() *PushCtx {
 // ctx mustn't be used after returning to the pool.
 func PutPushCtx(ctx *PushCtx) {
 	ctx.Reset()
-	pushCtxPool.Put(ctx)
+	select {
+	case pushCtxPoolCh <- ctx:
+	default:
+		pushCtxPool.Put(ctx)
+	}
 }
 
 var pushCtxPool sync.Pool
+var pushCtxPoolCh = make(chan *PushCtx, cgroup.AvailableCPUs())

app/vmagent/csvimport/request_handler.go

@@ -6,10 +6,10 @@ import (
 	"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/prompb"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/csvimport"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
+	parserCommon "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
+	parser "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/csvimport"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/csvimport/stream"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/tenantmetrics"
 	"github.com/VictoriaMetrics/metrics"
 )
@@ -22,16 +22,16 @@ var (
 
 // InsertHandler processes csv data from req.
 func InsertHandler(at *auth.Token, req *http.Request) error {
-	extraLabels, err := protoparserutil.GetExtraLabels(req)
+	extraLabels, err := parserCommon.GetExtraLabels(req)
 	if err != nil {
 		return err
 	}
-	return stream.Parse(req, func(rows []csvimport.Row) error {
+	return stream.Parse(req, func(rows []parser.Row) error {
 		return insertRows(at, rows, extraLabels)
 	})
 }
 
-func insertRows(at *auth.Token, rows []csvimport.Row, extraLabels []prompb.Label) error {
+func insertRows(at *auth.Token, rows []parser.Row, extraLabels []prompbmarshal.Label) error {
 	ctx := common.GetPushCtx()
 	defer common.PutPushCtx(ctx)
 
@@ -41,23 +41,23 @@ func insertRows(at *auth.Token, rows []csvimport.Row, extraLabels []prompb.Label
 	for i := range rows {
 		r := &rows[i]
 		labelsLen := len(labels)
-		labels = append(labels, prompb.Label{
+		labels = append(labels, prompbmarshal.Label{
 			Name:  "__name__",
 			Value: r.Metric,
 		})
 		for j := range r.Tags {
 			tag := &r.Tags[j]
-			labels = append(labels, prompb.Label{
+			labels = append(labels, prompbmarshal.Label{
 				Name:  tag.Key,
 				Value: tag.Value,
 			})
 		}
 		labels = append(labels, extraLabels...)
-		samples = append(samples, prompb.Sample{
+		samples = append(samples, prompbmarshal.Sample{
 			Value:     r.Value,
 			Timestamp: r.Timestamp,
 		})
-		tssDst = append(tssDst, prompb.TimeSeries{
+		tssDst = append(tssDst, prompbmarshal.TimeSeries{
 			Labels:  labels[labelsLen:],
 			Samples: samples[len(samples)-1:],
 		})
@@ -65,9 +65,7 @@ func insertRows(at *auth.Token, rows []csvimport.Row, extraLabels []prompb.Label
 	ctx.WriteRequest.Timeseries = tssDst
 	ctx.Labels = labels
 	ctx.Samples = samples
-	if !remotewrite.TryPush(at, &ctx.WriteRequest) {
-		return remotewrite.ErrQueueFullHTTPRetry
-	}
+	remotewrite.Push(at, &ctx.WriteRequest)
 	rowsInserted.Add(len(rows))
 	if at != nil {
 		rowsTenantInserted.Get(at).Add(len(rows))

app/vmagent/datadog/request_handler.go

@@ -0,0 +1,98 @@
+package datadog
+
+import (
+	"net/http"
+
+	"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"
+	parserCommon "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
+	parser "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/datadog"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/datadog/stream"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/tenantmetrics"
+	"github.com/VictoriaMetrics/metrics"
+)
+
+var (
+	rowsInserted       = metrics.NewCounter(`vmagent_rows_inserted_total{type="datadog"}`)
+	rowsTenantInserted = tenantmetrics.NewCounterMap(`vmagent_tenant_inserted_rows_total{type="datadog"}`)
+	rowsPerInsert      = metrics.NewHistogram(`vmagent_rows_per_insert{type="datadog"}`)
+)
+
+// InsertHandlerForHTTP processes remote write for DataDog POST /api/v1/series request.
+//
+// See https://docs.datadoghq.com/api/latest/metrics/#submit-metrics
+func InsertHandlerForHTTP(at *auth.Token, req *http.Request) error {
+	extraLabels, err := parserCommon.GetExtraLabels(req)
+	if err != nil {
+		return err
+	}
+	ce := req.Header.Get("Content-Encoding")
+	return stream.Parse(req.Body, ce, func(series []parser.Series) error {
+		return insertRows(at, series, extraLabels)
+	})
+}
+
+func insertRows(at *auth.Token, series []parser.Series, extraLabels []prompbmarshal.Label) error {
+	ctx := common.GetPushCtx()
+	defer common.PutPushCtx(ctx)
+
+	rowsTotal := 0
+	tssDst := ctx.WriteRequest.Timeseries[:0]
+	labels := ctx.Labels[:0]
+	samples := ctx.Samples[:0]
+	for i := range series {
+		ss := &series[i]
+		rowsTotal += len(ss.Points)
+		labelsLen := len(labels)
+		labels = append(labels, prompbmarshal.Label{
+			Name:  "__name__",
+			Value: ss.Metric,
+		})
+		if ss.Host != "" {
+			labels = append(labels, prompbmarshal.Label{
+				Name:  "host",
+				Value: ss.Host,
+			})
+		}
+		if ss.Device != "" {
+			labels = append(labels, prompbmarshal.Label{
+				Name:  "device",
+				Value: ss.Device,
+			})
+		}
+		for _, tag := range ss.Tags {
+			name, value := parser.SplitTag(tag)
+			if name == "host" {
+				name = "exported_host"
+			}
+			labels = append(labels, prompbmarshal.Label{
+				Name:  name,
+				Value: value,
+			})
+		}
+		labels = append(labels, extraLabels...)
+		samplesLen := len(samples)
+		for _, pt := range ss.Points {
+			samples = append(samples, prompbmarshal.Sample{
+				Timestamp: pt.Timestamp(),
+				Value:     pt.Value(),
+			})
+		}
+		tssDst = append(tssDst, prompbmarshal.TimeSeries{
+			Labels:  labels[labelsLen:],
+			Samples: samples[samplesLen:],
+		})
+	}
+	ctx.WriteRequest.Timeseries = tssDst
+	ctx.Labels = labels
+	ctx.Samples = samples
+	remotewrite.Push(at, &ctx.WriteRequest)
+	rowsInserted.Add(rowsTotal)
+	if at != nil {
+		rowsTenantInserted.Get(at).Add(rowsTotal)
+	}
+	rowsPerInsert.Update(float64(rowsTotal))
+	return nil
+}

app/vmagent/datadogsketches/request_handler.go

@@ -1,95 +0,0 @@
-package datadogsketches
-
-import (
-	"net/http"
-
-	"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/prompb"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/datadogsketches"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/datadogsketches/stream"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/datadogutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/tenantmetrics"
-	"github.com/VictoriaMetrics/metrics"
-)
-
-var (
-	rowsInserted       = metrics.NewCounter(`vmagent_rows_inserted_total{type="datadogsketches"}`)
-	rowsTenantInserted = tenantmetrics.NewCounterMap(`vmagent_tenant_inserted_rows_total{type="datadogsketches"}`)
-	rowsPerInsert      = metrics.NewHistogram(`vmagent_rows_per_insert{type="datadogsketches"}`)
-)
-
-// InsertHandlerForHTTP processes remote write for DataDog POST /api/beta/sketches request.
-func InsertHandlerForHTTP(at *auth.Token, req *http.Request) error {
-	extraLabels, err := protoparserutil.GetExtraLabels(req)
-	if err != nil {
-		return err
-	}
-	ce := req.Header.Get("Content-Encoding")
-	return stream.Parse(req.Body, ce, func(sketches []*datadogsketches.Sketch) error {
-		return insertRows(at, sketches, extraLabels)
-	})
-}
-
-func insertRows(at *auth.Token, sketches []*datadogsketches.Sketch, extraLabels []prompb.Label) error {
-	ctx := common.GetPushCtx()
-	defer common.PutPushCtx(ctx)
-
-	rowsTotal := 0
-	tssDst := ctx.WriteRequest.Timeseries[:0]
-	labels := ctx.Labels[:0]
-	samples := ctx.Samples[:0]
-	for _, sketch := range sketches {
-		ms := sketch.ToSummary()
-		for _, m := range ms {
-			labelsLen := len(labels)
-			labels = append(labels, prompb.Label{
-				Name:  "__name__",
-				Value: m.Name,
-			})
-			for _, label := range m.Labels {
-				labels = append(labels, prompb.Label{
-					Name:  label.Name,
-					Value: label.Value,
-				})
-			}
-			for _, tag := range sketch.Tags {
-				name, value := datadogutil.SplitTag(tag)
-				if name == "host" {
-					name = "exported_host"
-				}
-				labels = append(labels, prompb.Label{
-					Name:  name,
-					Value: value,
-				})
-			}
-			labels = append(labels, extraLabels...)
-			samplesLen := len(samples)
-			for _, p := range m.Points {
-				samples = append(samples, prompb.Sample{
-					Timestamp: p.Timestamp,
-					Value:     p.Value,
-				})
-			}
-			rowsTotal += len(m.Points)
-			tssDst = append(tssDst, prompb.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(rowsTotal)
-	if at != nil {
-		rowsTenantInserted.Get(at).Add(rowsTotal)
-	}
-	rowsPerInsert.Update(float64(rowsTotal))
-	return nil
-}

app/vmagent/datadogv1/request_handler.go

@@ -1,99 +0,0 @@
-package datadogv1
-
-import (
-	"net/http"
-
-	"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/prompb"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/datadogutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/datadogv1"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/datadogv1/stream"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/tenantmetrics"
-	"github.com/VictoriaMetrics/metrics"
-)
-
-var (
-	rowsInserted       = metrics.NewCounter(`vmagent_rows_inserted_total{type="datadogv1"}`)
-	rowsTenantInserted = tenantmetrics.NewCounterMap(`vmagent_tenant_inserted_rows_total{type="datadogv1"}`)
-	rowsPerInsert      = metrics.NewHistogram(`vmagent_rows_per_insert{type="datadogv1"}`)
-)
-
-// InsertHandlerForHTTP processes remote write for DataDog POST /api/v1/series request.
-func InsertHandlerForHTTP(at *auth.Token, req *http.Request) error {
-	extraLabels, err := protoparserutil.GetExtraLabels(req)
-	if err != nil {
-		return err
-	}
-	ce := req.Header.Get("Content-Encoding")
-	return stream.Parse(req.Body, ce, func(series []datadogv1.Series) error {
-		return insertRows(at, series, extraLabels)
-	})
-}
-
-func insertRows(at *auth.Token, series []datadogv1.Series, extraLabels []prompb.Label) error {
-	ctx := common.GetPushCtx()
-	defer common.PutPushCtx(ctx)
-
-	rowsTotal := 0
-	tssDst := ctx.WriteRequest.Timeseries[:0]
-	labels := ctx.Labels[:0]
-	samples := ctx.Samples[:0]
-	for i := range series {
-		ss := &series[i]
-		rowsTotal += len(ss.Points)
-		labelsLen := len(labels)
-		labels = append(labels, prompb.Label{
-			Name:  "__name__",
-			Value: ss.Metric,
-		})
-		if ss.Host != "" {
-			labels = append(labels, prompb.Label{
-				Name:  "host",
-				Value: ss.Host,
-			})
-		}
-		if ss.Device != "" {
-			labels = append(labels, prompb.Label{
-				Name:  "device",
-				Value: ss.Device,
-			})
-		}
-		for _, tag := range ss.Tags {
-			name, value := datadogutil.SplitTag(tag)
-			if name == "host" {
-				name = "exported_host"
-			}
-			labels = append(labels, prompb.Label{
-				Name:  name,
-				Value: value,
-			})
-		}
-		labels = append(labels, extraLabels...)
-		samplesLen := len(samples)
-		for _, pt := range ss.Points {
-			samples = append(samples, prompb.Sample{
-				Timestamp: pt.Timestamp(),
-				Value:     pt.Value(),
-			})
-		}
-		tssDst = append(tssDst, prompb.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(rowsTotal)
-	if at != nil {
-		rowsTenantInserted.Get(at).Add(rowsTotal)
-	}
-	rowsPerInsert.Update(float64(rowsTotal))
-	return nil
-}

app/vmagent/datadogv2/request_handler.go

@@ -1,102 +0,0 @@
-package datadogv2
-
-import (
-	"net/http"
-
-	"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/prompb"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/datadogutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/datadogv2"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/datadogv2/stream"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/tenantmetrics"
-	"github.com/VictoriaMetrics/metrics"
-)
-
-var (
-	rowsInserted       = metrics.NewCounter(`vmagent_rows_inserted_total{type="datadogv2"}`)
-	rowsTenantInserted = tenantmetrics.NewCounterMap(`vmagent_tenant_inserted_rows_total{type="datadogv2"}`)
-	rowsPerInsert      = metrics.NewHistogram(`vmagent_rows_per_insert{type="datadogv2"}`)
-)
-
-// InsertHandlerForHTTP processes remote write for DataDog POST /api/v2/series request.
-//
-// See https://docs.datadoghq.com/api/latest/metrics/#submit-metrics
-func InsertHandlerForHTTP(at *auth.Token, req *http.Request) error {
-	extraLabels, err := protoparserutil.GetExtraLabels(req)
-	if err != nil {
-		return err
-	}
-	ct := req.Header.Get("Content-Type")
-	ce := req.Header.Get("Content-Encoding")
-	return stream.Parse(req.Body, ce, ct, func(series []datadogv2.Series) error {
-		return insertRows(at, series, extraLabels)
-	})
-}
-
-func insertRows(at *auth.Token, series []datadogv2.Series, extraLabels []prompb.Label) error {
-	ctx := common.GetPushCtx()
-	defer common.PutPushCtx(ctx)
-
-	rowsTotal := 0
-	tssDst := ctx.WriteRequest.Timeseries[:0]
-	labels := ctx.Labels[:0]
-	samples := ctx.Samples[:0]
-	for i := range series {
-		ss := &series[i]
-		rowsTotal += len(ss.Points)
-		labelsLen := len(labels)
-		labels = append(labels, prompb.Label{
-			Name:  "__name__",
-			Value: ss.Metric,
-		})
-		for _, rs := range ss.Resources {
-			labels = append(labels, prompb.Label{
-				Name:  rs.Type,
-				Value: rs.Name,
-			})
-		}
-		if ss.SourceTypeName != "" {
-			labels = append(labels, prompb.Label{
-				Name:  "source_type_name",
-				Value: ss.SourceTypeName,
-			})
-		}
-		for _, tag := range ss.Tags {
-			name, value := datadogutil.SplitTag(tag)
-			if name == "host" {
-				name = "exported_host"
-			}
-			labels = append(labels, prompb.Label{
-				Name:  name,
-				Value: value,
-			})
-		}
-		labels = append(labels, extraLabels...)
-		samplesLen := len(samples)
-		for _, pt := range ss.Points {
-			samples = append(samples, prompb.Sample{
-				Timestamp: pt.Timestamp * 1000,
-				Value:     pt.Value,
-			})
-		}
-		tssDst = append(tssDst, prompb.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(rowsTotal)
-	if at != nil {
-		rowsTenantInserted.Get(at).Add(rowsTotal)
-	}
-	rowsPerInsert.Update(float64(rowsTotal))
-	return nil
-}

app/vmagent/deployment/Dockerfile

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

app/vmagent/graphite/request_handler.go

@@ -5,8 +5,7 @@ import (
 
 	"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/prompb"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
 	parser "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/graphite"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/graphite/stream"
 	"github.com/VictoriaMetrics/metrics"
@@ -21,12 +20,10 @@ var (
 //
 // See https://graphite.readthedocs.io/en/latest/feeding-carbon.html#the-plaintext-protocol
 func InsertHandler(r io.Reader) error {
-	return stream.Parse(r, "", func(rows []parser.Row) error {
-		return insertRows(nil, rows)
-	})
+	return stream.Parse(r, insertRows)
 }
 
-func insertRows(at *auth.Token, rows []parser.Row) error {
+func insertRows(rows []parser.Row) error {
 	ctx := common.GetPushCtx()
 	defer common.PutPushCtx(ctx)
 
@@ -36,22 +33,22 @@ func insertRows(at *auth.Token, rows []parser.Row) error {
 	for i := range rows {
 		r := &rows[i]
 		labelsLen := len(labels)
-		labels = append(labels, prompb.Label{
+		labels = append(labels, prompbmarshal.Label{
 			Name:  "__name__",
 			Value: r.Metric,
 		})
 		for j := range r.Tags {
 			tag := &r.Tags[j]
-			labels = append(labels, prompb.Label{
+			labels = append(labels, prompbmarshal.Label{
 				Name:  tag.Key,
 				Value: tag.Value,
 			})
 		}
-		samples = append(samples, prompb.Sample{
+		samples = append(samples, prompbmarshal.Sample{
 			Value:     r.Value,
 			Timestamp: r.Timestamp,
 		})
-		tssDst = append(tssDst, prompb.TimeSeries{
+		tssDst = append(tssDst, prompbmarshal.TimeSeries{
 			Labels:  labels[labelsLen:],
 			Samples: samples[len(samples)-1:],
 		})
@@ -59,9 +56,7 @@ func insertRows(at *auth.Token, rows []parser.Row) error {
 	ctx.WriteRequest.Timeseries = tssDst
 	ctx.Labels = labels
 	ctx.Samples = samples
-	if !remotewrite.TryPush(at, &ctx.WriteRequest) {
-		return remotewrite.ErrQueueFullHTTPRetry
-	}
+	remotewrite.Push(nil, &ctx.WriteRequest)
 	rowsInserted.Add(len(rows))
 	rowsPerInsert.Update(float64(len(rows)))
 	return nil

app/vmagent/influx/request_handler.go

@@ -10,11 +10,12 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/remotewrite"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/auth"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/cgroup"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promrelabel"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/influx"
+	parserCommon "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
+	parser "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/influx"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/influx/stream"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/tenantmetrics"
 	"github.com/VictoriaMetrics/metrics"
 )
@@ -35,9 +36,9 @@ var (
 // InsertHandlerForReader processes remote write for influx line protocol.
 //
 // See https://github.com/influxdata/telegraf/tree/master/plugins/inputs/socket_listener/
-func InsertHandlerForReader(at *auth.Token, r io.Reader, encoding string) error {
-	return stream.Parse(r, encoding, true, "", "", func(db string, rows []influx.Row) error {
-		return insertRows(at, db, rows, nil)
+func InsertHandlerForReader(r io.Reader, isGzipped bool) error {
+	return stream.Parse(r, isGzipped, "", "", func(db string, rows []parser.Row) error {
+		return insertRows(nil, db, rows, nil)
 	})
 }
 
@@ -45,22 +46,21 @@ func InsertHandlerForReader(at *auth.Token, r io.Reader, encoding string) error
 //
 // See https://github.com/influxdata/influxdb/blob/4cbdc197b8117fee648d62e2e5be75c6575352f0/tsdb/README.md
 func InsertHandlerForHTTP(at *auth.Token, req *http.Request) error {
-	extraLabels, err := protoparserutil.GetExtraLabels(req)
+	extraLabels, err := parserCommon.GetExtraLabels(req)
 	if err != nil {
 		return err
 	}
+	isGzipped := req.Header.Get("Content-Encoding") == "gzip"
 	q := req.URL.Query()
 	precision := q.Get("precision")
 	// Read db tag from https://docs.influxdata.com/influxdb/v1.7/tools/api/#write-http-endpoint
 	db := q.Get("db")
-	encoding := req.Header.Get("Content-Encoding")
-	isStreamMode := req.Header.Get("Stream-Mode") == "1"
-	return stream.Parse(req.Body, encoding, isStreamMode, precision, db, func(db string, rows []influx.Row) error {
+	return stream.Parse(req.Body, isGzipped, precision, db, func(db string, rows []parser.Row) error {
 		return insertRows(at, db, rows, extraLabels)
 	})
 }
 
-func insertRows(at *auth.Token, db string, rows []influx.Row, extraLabels []prompb.Label) error {
+func insertRows(at *auth.Token, db string, rows []parser.Row, extraLabels []prompbmarshal.Label) error {
 	ctx := getPushCtx()
 	defer putPushCtx(ctx)
 
@@ -80,13 +80,13 @@ func insertRows(at *auth.Token, db string, rows []influx.Row, extraLabels []prom
 			if tag.Key == *dbLabel {
 				hasDBKey = true
 			}
-			commonLabels = append(commonLabels, prompb.Label{
+			commonLabels = append(commonLabels, prompbmarshal.Label{
 				Name:  tag.Key,
 				Value: tag.Value,
 			})
 		}
 		if len(db) > 0 && !hasDBKey {
-			commonLabels = append(commonLabels, prompb.Label{
+			commonLabels = append(commonLabels, prompbmarshal.Label{
 				Name:  *dbLabel,
 				Value: db,
 			})
@@ -110,16 +110,16 @@ func insertRows(at *auth.Token, db string, rows []influx.Row, extraLabels []prom
 			}
 			metricGroup := bytesutil.ToUnsafeString(buf[bufLen:])
 			labelsLen := len(labels)
-			labels = append(labels, prompb.Label{
+			labels = append(labels, prompbmarshal.Label{
 				Name:  "__name__",
 				Value: metricGroup,
 			})
 			labels = append(labels, commonLabels...)
-			samples = append(samples, prompb.Sample{
+			samples = append(samples, prompbmarshal.Sample{
 				Timestamp: r.Timestamp,
 				Value:     f.Value,
 			})
-			tssDst = append(tssDst, prompb.TimeSeries{
+			tssDst = append(tssDst, prompbmarshal.TimeSeries{
 				Labels:  labels[labelsLen:],
 				Samples: samples[len(samples)-1:],
 			})
@@ -130,9 +130,7 @@ func insertRows(at *auth.Token, db string, rows []influx.Row, extraLabels []prom
 	ctx.ctx.Labels = labels
 	ctx.ctx.Samples = samples
 	ctx.commonLabels = commonLabels
-	if !remotewrite.TryPush(at, &ctx.ctx.WriteRequest) {
-		return remotewrite.ErrQueueFullHTTPRetry
-	}
+	remotewrite.Push(at, &ctx.ctx.WriteRequest)
 	rowsInserted.Add(rowsTotal)
 	if at != nil {
 		rowsTenantInserted.Get(at).Add(rowsTotal)
@@ -144,7 +142,7 @@ func insertRows(at *auth.Token, db string, rows []influx.Row, extraLabels []prom
 
 type pushCtx struct {
 	ctx            common.PushCtx
-	commonLabels   []prompb.Label
+	commonLabels   []prompbmarshal.Label
 	metricGroupBuf []byte
 	buf            []byte
 }
@@ -160,15 +158,25 @@ func (ctx *pushCtx) reset() {
 }
 
 func getPushCtx() *pushCtx {
-	if v := pushCtxPool.Get(); v != nil {
-		return v.(*pushCtx)
+	select {
+	case ctx := <-pushCtxPoolCh:
+		return ctx
+	default:
+		if v := pushCtxPool.Get(); v != nil {
+			return v.(*pushCtx)
+		}
+		return &pushCtx{}
 	}
-	return &pushCtx{}
 }
 
 func putPushCtx(ctx *pushCtx) {
 	ctx.reset()
-	pushCtxPool.Put(ctx)
+	select {
+	case pushCtxPoolCh <- ctx:
+	default:
+		pushCtxPool.Put(ctx)
+	}
 }
 
 var pushCtxPool sync.Pool
+var pushCtxPoolCh = make(chan *pushCtx, cgroup.AvailableCPUs())

app/vmagent/main.go

@@ -8,19 +8,14 @@ import (
 	"net/http"
 	"os"
 	"strings"
+	"sync/atomic"
 	"time"
 
-	"github.com/VictoriaMetrics/metrics"
-
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/csvimport"
-	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/datadogsketches"
-	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/datadogv1"
-	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/datadogv2"
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/datadog"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/graphite"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/influx"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/native"
-	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/newrelic"
-	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/opentelemetry"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/opentsdb"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/opentsdbhttp"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/prometheusimport"
@@ -30,11 +25,10 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/auth"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/buildinfo"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/cgroup"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/envflag"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/influxutil"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/influxutils"
 	graphiteserver "github.com/VictoriaMetrics/VictoriaMetrics/lib/ingestserver/graphite"
 	influxserver "github.com/VictoriaMetrics/VictoriaMetrics/lib/ingestserver/influx"
 	opentsdbserver "github.com/VictoriaMetrics/VictoriaMetrics/lib/ingestserver/opentsdb"
@@ -42,18 +36,16 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/procutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promscrape"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentelemetry/firehose"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/pushmetrics"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/stringsutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/timeserieslimits"
+	"github.com/VictoriaMetrics/metrics"
 )
 
 var (
-	httpListenAddrs = flagutil.NewArrayString("httpListenAddr", "TCP address to listen for incoming http requests. "+
+	httpListenAddr = flag.String("httpListenAddr", ":8429", "TCP address to listen for http connections. "+
 		"Set this flag to empty value in order to disable listening on any port. This mode may be useful for running multiple vmagent instances on the same server. "+
-		"Note that /targets and /metrics pages aren't available if -httpListenAddr=''. See also -tls and -httpListenAddr.useProxyProtocol")
-	useProxyProtocol = flagutil.NewArrayBool("httpListenAddr.useProxyProtocol", "Whether to use proxy protocol for connections accepted at the corresponding -httpListenAddr . "+
+		"Note that /targets and /metrics pages aren't available if -httpListenAddr=''. See also -httpListenAddr.useProxyProtocol")
+	useProxyProtocol = flag.Bool("httpListenAddr.useProxyProtocol", false, "Whether to use proxy protocol for connections accepted at -httpListenAddr . "+
 		"See https://www.haproxy.org/download/1.8/doc/proxy-protocol.txt . "+
 		"With enabled proxy protocol http server cannot serve regular /metrics endpoint. Use -pushmetrics.url for metrics pushing")
 	influxListenAddr = flag.String("influxListenAddr", "", "TCP and UDP address to listen for InfluxDB line protocol data. Usually :8089 must be set. Doesn't work if empty. "+
@@ -74,14 +66,10 @@ 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.*")
-	reloadAuthKey = flagutil.NewPassword("reloadAuthKey", "Auth key for /-/reload http endpoint. It must be passed via authKey query arg. It overrides -httpAuth.*")
+	configAuthKey = flag.String("configAuthKey", "", "Authorization key for accessing /config page. It must be passed via authKey query arg")
 	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")
-	maxLabelsPerTimeseries = flag.Int("maxLabelsPerTimeseries", 0, "The maximum number of labels per time series to be accepted. Series with superfluous labels are ignored. In this case the vm_rows_ignored_total{reason=\"too_many_labels\"} metric at /metrics page is incremented")
-	maxLabelNameLen        = flag.Int("maxLabelNameLen", 0, "The maximum length of label names in the accepted time series. Series with longer label name are ignored. In this case the vm_rows_ignored_total{reason=\"too_long_label_name\"} metric at /metrics page is incremented")
-	maxLabelValueLen       = flag.Int("maxLabelValueLen", 0, "The maximum length of label values in the accepted time series. Series with longer label value are ignored. In this case the vm_rows_ignored_total{reason=\"too_long_label_value\"} metric at /metrics page is incremented")
 )
 
 var (
@@ -98,15 +86,6 @@ var (
 )
 
 func main() {
-	// vmagent is optimized for reduced memory allocations,
-	// so it can run with the reduced GOGC in order to reduce the used memory,
-	// while keeping CPU usage spent in GC at low levels.
-	//
-	// Some workloads may need increased GOGC values. Then such values can be set via GOGC environment variable.
-	// It is recommended increasing GOGC if go_memstats_gc_cpu_fraction metric exposed at /metrics page
-	// exceeds 0.05 for extended periods of time.
-	cgroup.SetGOGC(50)
-
 	// Write flags and help message to stdout, since it is easier to grep or pipe.
 	flag.CommandLine.SetOutput(os.Stdout)
 	flag.Usage = usage
@@ -114,7 +93,7 @@ func main() {
 	remotewrite.InitSecretFlags()
 	buildinfo.Init()
 	logger.Init()
-	timeserieslimits.Init(*maxLabelsPerTimeseries, *maxLabelNameLen, *maxLabelValueLen)
+	pushmetrics.Init()
 
 	if promscrape.IsDryRun() {
 		if err := promscrape.CheckConfig(); err != nil {
@@ -131,24 +110,19 @@ func main() {
 			logger.Fatalf("error when checking relabel configs: %s", err)
 		}
 		if err := remotewrite.CheckStreamAggrConfigs(); err != nil {
-			logger.Fatalf("error when checking -streamAggr.config and -remoteWrite.streamAggr.config: %s", err)
+			logger.Fatalf("error when checking -remoteWrite.streamAggr.config: %s", err)
 		}
 		logger.Infof("all the configs are ok; exiting with 0 status code")
 		return
 	}
 
-	listenAddrs := *httpListenAddrs
-	if len(listenAddrs) == 0 {
-		listenAddrs = []string{":8429"}
-	}
-	logger.Infof("starting vmagent at %q...", listenAddrs)
+	logger.Infof("starting vmagent at %q...", *httpListenAddr)
 	startTime := time.Now()
-	remotewrite.StartIngestionRateLimiter()
 	remotewrite.Init()
-	protoparserutil.StartUnmarshalWorkers()
+	common.StartUnmarshalWorkers()
 	if len(*influxListenAddr) > 0 {
 		influxServer = influxserver.MustStart(*influxListenAddr, *influxUseProxyProtocol, func(r io.Reader) error {
-			return influx.InsertHandlerForReader(nil, r, "")
+			return influx.InsertHandlerForReader(r, false)
 		})
 	}
 	if len(*graphiteListenAddr) > 0 {
@@ -163,25 +137,24 @@ func main() {
 		opentsdbhttpServer = opentsdbhttpserver.MustStart(*opentsdbHTTPListenAddr, *opentsdbHTTPUseProxyProtocol, httpInsertHandler)
 	}
 
-	promscrape.Init(remotewrite.PushDropSamplesOnFailure)
+	promscrape.Init(remotewrite.Push)
 
-	go httpserver.Serve(listenAddrs, requestHandler, httpserver.ServeOptions{
-		UseProxyProtocol: useProxyProtocol,
-	})
+	if len(*httpListenAddr) > 0 {
+		go httpserver.Serve(*httpListenAddr, *useProxyProtocol, requestHandler)
+	}
 	logger.Infof("started vmagent in %.3f seconds", time.Since(startTime).Seconds())
 
-	pushmetrics.Init()
 	sig := procutil.WaitForSigterm()
 	logger.Infof("received signal %s", sig)
-	remotewrite.StopIngestionRateLimiter()
-	pushmetrics.Stop()
 
 	startTime = time.Now()
-	logger.Infof("gracefully shutting down webservice at %q", listenAddrs)
-	if err := httpserver.Stop(listenAddrs); err != nil {
-		logger.Fatalf("cannot stop the webservice: %s", err)
+	if len(*httpListenAddr) > 0 {
+		logger.Infof("gracefully shutting down webservice at %q", *httpListenAddr)
+		if err := httpserver.Stop(*httpListenAddr); err != nil {
+			logger.Fatalf("cannot stop the webservice: %s", err)
+		}
+		logger.Infof("successfully shut down the webservice in %.3f seconds", time.Since(startTime).Seconds())
 	}
-	logger.Infof("successfully shut down the webservice in %.3f seconds", time.Since(startTime).Seconds())
 
 	promscrape.Stop()
 
@@ -197,7 +170,7 @@ func main() {
 	if len(*opentsdbHTTPListenAddr) > 0 {
 		opentsdbhttpServer.MustStop()
 	}
-	protoparserutil.StopUnmarshalWorkers()
+	common.StopUnmarshalWorkers()
 	remotewrite.Stop()
 
 	logger.Infof("successfully stopped vmagent in %.3f seconds", time.Since(startTime).Seconds())
@@ -206,7 +179,7 @@ func main() {
 func getOpenTSDBHTTPInsertHandler() func(req *http.Request) error {
 	if !remotewrite.MultitenancyEnabled() {
 		return func(req *http.Request) error {
-			path := strings.ReplaceAll(req.URL.Path, "//", "/")
+			path := strings.Replace(req.URL.Path, "//", "/", -1)
 			if path != "/api/put" {
 				return fmt.Errorf("unsupported path requested: %q; expecting '/api/put'", path)
 			}
@@ -214,7 +187,7 @@ func getOpenTSDBHTTPInsertHandler() func(req *http.Request) error {
 		}
 	}
 	return func(req *http.Request) error {
-		path := strings.ReplaceAll(req.URL.Path, "//", "/")
+		path := strings.Replace(req.URL.Path, "//", "/", -1)
 		at, err := getAuthTokenFromPath(path)
 		if err != nil {
 			return fmt.Errorf("cannot obtain auth token from path %q: %w", path, err)
@@ -234,7 +207,7 @@ func getAuthTokenFromPath(path string) (*auth.Token, error) {
 	if p.Suffix != "opentsdb/api/put" {
 		return nil, fmt.Errorf("unsupported path requested: %q; expecting 'opentsdb/api/put'", p.Suffix)
 	}
-	return auth.NewTokenPossibleMultitenant(p.AuthToken)
+	return auth.NewToken(p.AuthToken)
 }
 
 func requestHandler(w http.ResponseWriter, r *http.Request) bool {
@@ -244,7 +217,7 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 		}
 		w.Header().Add("Content-Type", "text/html; charset=utf-8")
 		fmt.Fprintf(w, "<h2>vmagent</h2>")
-		fmt.Fprintf(w, "See docs at <a href='https://docs.victoriametrics.com/victoriametrics/vmagent/'>https://docs.victoriametrics.com/victoriametrics/vmagent/</a></br>")
+		fmt.Fprintf(w, "See docs at <a href='https://docs.victoriametrics.com/vmagent.html'>https://docs.victoriametrics.com/vmagent.html</a></br>")
 		fmt.Fprintf(w, "Useful endpoints:</br>")
 		httpserver.WriteAPIHelp(w, [][2]string{
 			{"targets", "status for discovered active targets"},
@@ -259,7 +232,7 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 		return true
 	}
 
-	path := strings.ReplaceAll(r.URL.Path, "//", "/")
+	path := strings.Replace(r.URL.Path, "//", "/", -1)
 	if strings.HasPrefix(path, "/prometheus/api/v1/import/prometheus") || strings.HasPrefix(path, "/api/v1/import/prometheus") {
 		prometheusimportRequests.Inc()
 		if err := prometheusimport.InsertHandler(nil, r); err != nil {
@@ -277,14 +250,14 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 		w.WriteHeader(statusCode)
 		return true
 	}
-	if strings.HasPrefix(path, "/datadog/") {
+	if strings.HasPrefix(path, "datadog/") {
 		// Trim suffix from paths starting from /datadog/ in order to support legacy DataDog agent.
 		// See https://github.com/VictoriaMetrics/VictoriaMetrics/pull/2670
 		path = strings.TrimSuffix(path, "/")
 	}
 	switch path {
-	case "/prometheus/api/v1/write", "/api/v1/write", "/api/v1/push", "/prometheus/api/v1/push":
-		if protoparserutil.HandleVMProtoServerHandshake(w, r) {
+	case "/prometheus/api/v1/write", "/api/v1/write":
+		if common.HandleVMProtoServerHandshake(w, r) {
 			return true
 		}
 		prometheusWriteRequests.Inc()
@@ -333,59 +306,12 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 		return true
 	case "/influx/query", "/query":
 		influxQueryRequests.Inc()
-		influxutil.WriteDatabaseNames(w)
-		return true
-	case "/influx/health":
-		influxHealthRequests.Inc()
-		influxutil.WriteHealthCheckResponse(w)
-		return true
-	case "/opentelemetry/api/v1/push", "/opentelemetry/v1/metrics":
-		opentelemetryPushRequests.Inc()
-		if err := opentelemetry.InsertHandler(nil, r); err != nil {
-			opentelemetryPushErrors.Inc()
-			httpserver.Errorf(w, r, "%s", err)
-			return true
-		}
-		firehose.WriteSuccessResponse(w, r)
-		return true
-	case "/newrelic":
-		newrelicCheckRequest.Inc()
-		w.Header().Set("Content-Type", "application/json")
-		w.WriteHeader(202)
-		fmt.Fprintf(w, `{"status":"ok"}`)
-		return true
-	case "/newrelic/inventory/deltas":
-		newrelicInventoryRequests.Inc()
-		w.Header().Set("Content-Type", "application/json")
-		w.WriteHeader(202)
-		fmt.Fprintf(w, `{"payload":{"version": 1, "state": {}, "reset": "false"}}`)
-		return true
-	case "/newrelic/infra/v2/metrics/events/bulk":
-		newrelicWriteRequests.Inc()
-		if err := newrelic.InsertHandlerForHTTP(nil, r); err != nil {
-			newrelicWriteErrors.Inc()
-			httpserver.Errorf(w, r, "%s", err)
-			return true
-		}
-		w.Header().Set("Content-Type", "application/json")
-		w.WriteHeader(202)
-		fmt.Fprintf(w, `{"status":"ok"}`)
+		influxutils.WriteDatabaseNames(w)
 		return true
 	case "/datadog/api/v1/series":
-		datadogv1WriteRequests.Inc()
-		if err := datadogv1.InsertHandlerForHTTP(nil, r); err != nil {
-			datadogv1WriteErrors.Inc()
-			httpserver.Errorf(w, r, "%s", err)
-			return true
-		}
-		w.Header().Set("Content-Type", "application/json")
-		w.WriteHeader(202)
-		fmt.Fprintf(w, `{"status":"ok"}`)
-		return true
-	case "/datadog/api/v2/series":
-		datadogv2WriteRequests.Inc()
-		if err := datadogv2.InsertHandlerForHTTP(nil, r); err != nil {
-			datadogv2WriteErrors.Inc()
+		datadogWriteRequests.Inc()
+		if err := datadog.InsertHandlerForHTTP(nil, r); err != nil {
+			datadogWriteErrors.Inc()
 			httpserver.Errorf(w, r, "%s", err)
 			return true
 		}
@@ -394,15 +320,6 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 		w.WriteHeader(202)
 		fmt.Fprintf(w, `{"status":"ok"}`)
 		return true
-	case "/datadog/api/beta/sketches":
-		datadogsketchesWriteRequests.Inc()
-		if err := datadogsketches.InsertHandlerForHTTP(nil, r); err != nil {
-			datadogsketchesWriteErrors.Inc()
-			httpserver.Errorf(w, r, "%s", err)
-			return true
-		}
-		w.WriteHeader(202)
-		return true
 	case "/datadog/api/v1/validate":
 		datadogValidateRequests.Inc()
 		// See https://docs.datadoghq.com/api/latest/authentication/#validate-api-key
@@ -445,10 +362,8 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 	case "/prometheus/api/v1/targets", "/api/v1/targets":
 		promscrapeAPIV1TargetsRequests.Inc()
 		w.Header().Set("Content-Type", "application/json")
-		// https://prometheus.io/docs/prometheus/latest/querying/api/#targets
 		state := r.FormValue("state")
-		scrapePool := r.FormValue("scrapePool")
-		promscrape.WriteAPIV1Targets(w, state, scrapePool)
+		promscrape.WriteAPIV1Targets(w, state)
 		return true
 	case "/prometheus/target_response", "/target_response":
 		promscrapeTargetResponseRequests.Inc()
@@ -459,7 +374,7 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
 		}
 		return true
 	case "/prometheus/config", "/config":
-		if !httpserver.CheckAuthFlag(w, r, configAuthKey) {
+		if !httpserver.CheckAuthFlag(w, r, *configAuthKey, "configAuthKey") {
 			return true
 		}
 		promscrapeConfigRequests.Inc()
@@ -468,25 +383,22 @@ 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) {
+		if !httpserver.CheckAuthFlag(w, r, *configAuthKey, "configAuthKey") {
 			return true
 		}
 		promscrapeStatusConfigRequests.Inc()
 		w.Header().Set("Content-Type", "application/json")
 		var bb bytesutil.ByteBuffer
 		promscrape.WriteConfigData(&bb)
-		fmt.Fprintf(w, `{"status":"success","data":{"yaml":%s}}`, stringsutil.JSONString(string(bb.B)))
+		fmt.Fprintf(w, `{"status":"success","data":{"yaml":%q}}`, bb.B)
 		return true
 	case "/prometheus/-/reload", "/-/reload":
-		if !httpserver.CheckAuthFlag(w, r, reloadAuthKey) {
-			return true
-		}
 		promscrapeConfigReloadRequests.Inc()
 		procutil.SelfSIGHUP()
 		w.WriteHeader(http.StatusOK)
 		return true
 	case "/ready":
-		if rdy := promscrape.PendingScrapeConfigs.Load(); rdy > 0 {
+		if rdy := atomic.LoadInt32(&promscrape.PendingScrapeConfigs); rdy > 0 {
 			errMsg := fmt.Sprintf("waiting for scrapes to init, left: %d", rdy)
 			http.Error(w, errMsg, http.StatusTooEarly)
 		} else {
@@ -517,7 +429,7 @@ func processMultitenantRequest(w http.ResponseWriter, r *http.Request, path stri
 		httpserver.Errorf(w, r, `unsupported multitenant prefix: %q; expected "insert"`, p.Prefix)
 		return true
 	}
-	at, err := auth.NewTokenPossibleMultitenant(p.AuthToken)
+	at, err := auth.NewToken(p.AuthToken)
 	if err != nil {
 		httpserver.Errorf(w, r, "cannot obtain auth token: %s", err)
 		return true
@@ -529,13 +441,7 @@ func processMultitenantRequest(w http.ResponseWriter, r *http.Request, path stri
 			httpserver.Errorf(w, r, "%s", err)
 			return true
 		}
-		statusCode := http.StatusNoContent
-		if strings.HasPrefix(p.Suffix, "prometheus/api/v1/import/prometheus/metrics/job/") {
-			// Return 200 status code for pushgateway requests.
-			// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/3636
-			statusCode = http.StatusOK
-		}
-		w.WriteHeader(statusCode)
+		w.WriteHeader(http.StatusNoContent)
 		return true
 	}
 	if strings.HasPrefix(p.Suffix, "datadog/") {
@@ -544,7 +450,7 @@ func processMultitenantRequest(w http.ResponseWriter, r *http.Request, path stri
 		p.Suffix = strings.TrimSuffix(p.Suffix, "/")
 	}
 	switch p.Suffix {
-	case "prometheus/", "prometheus", "prometheus/api/v1/write", "prometheus/api/v1/push":
+	case "prometheus/", "prometheus", "prometheus/api/v1/write":
 		prometheusWriteRequests.Inc()
 		if err := promremotewrite.InsertHandler(at, r); err != nil {
 			prometheusWriteErrors.Inc()
@@ -591,58 +497,12 @@ func processMultitenantRequest(w http.ResponseWriter, r *http.Request, path stri
 		return true
 	case "influx/query":
 		influxQueryRequests.Inc()
-		influxutil.WriteDatabaseNames(w)
-		return true
-	case "influx/health":
-		influxHealthRequests.Inc()
-		influxutil.WriteHealthCheckResponse(w)
-		return true
-	case "opentelemetry/api/v1/push", "opentelemetry/v1/metrics":
-		opentelemetryPushRequests.Inc()
-		if err := opentelemetry.InsertHandler(at, r); err != nil {
-			opentelemetryPushErrors.Inc()
-			httpserver.Errorf(w, r, "%s", err)
-			return true
-		}
-		firehose.WriteSuccessResponse(w, r)
-		return true
-	case "newrelic":
-		newrelicCheckRequest.Inc()
-		w.Header().Set("Content-Type", "application/json")
-		w.WriteHeader(202)
-		fmt.Fprintf(w, `{"status":"ok"}`)
-		return true
-	case "newrelic/inventory/deltas":
-		newrelicInventoryRequests.Inc()
-		w.Header().Set("Content-Type", "application/json")
-		w.WriteHeader(202)
-		fmt.Fprintf(w, `{"payload":{"version": 1, "state": {}, "reset": "false"}}`)
-		return true
-	case "newrelic/infra/v2/metrics/events/bulk":
-		newrelicWriteRequests.Inc()
-		if err := newrelic.InsertHandlerForHTTP(at, r); err != nil {
-			newrelicWriteErrors.Inc()
-			httpserver.Errorf(w, r, "%s", err)
-			return true
-		}
-		w.Header().Set("Content-Type", "application/json")
-		w.WriteHeader(202)
-		fmt.Fprintf(w, `{"status":"ok"}`)
+		influxutils.WriteDatabaseNames(w)
 		return true
 	case "datadog/api/v1/series":
-		datadogv1WriteRequests.Inc()
-		if err := datadogv1.InsertHandlerForHTTP(at, r); err != nil {
-			datadogv1WriteErrors.Inc()
-			httpserver.Errorf(w, r, "%s", err)
-			return true
-		}
-		w.WriteHeader(202)
-		fmt.Fprintf(w, `{"status":"ok"}`)
-		return true
-	case "datadog/api/v2/series":
-		datadogv2WriteRequests.Inc()
-		if err := datadogv2.InsertHandlerForHTTP(at, r); err != nil {
-			datadogv2WriteErrors.Inc()
+		datadogWriteRequests.Inc()
+		if err := datadog.InsertHandlerForHTTP(at, r); err != nil {
+			datadogWriteErrors.Inc()
 			httpserver.Errorf(w, r, "%s", err)
 			return true
 		}
@@ -650,15 +510,6 @@ func processMultitenantRequest(w http.ResponseWriter, r *http.Request, path stri
 		w.WriteHeader(202)
 		fmt.Fprintf(w, `{"status":"ok"}`)
 		return true
-	case "datadog/api/beta/sketches":
-		datadogsketchesWriteRequests.Inc()
-		if err := datadogsketches.InsertHandlerForHTTP(at, r); err != nil {
-			datadogsketchesWriteErrors.Inc()
-			httpserver.Errorf(w, r, "%s", err)
-			return true
-		}
-		w.WriteHeader(202)
-		return true
 	case "datadog/api/v1/validate":
 		datadogValidateRequests.Inc()
 		// See https://docs.datadoghq.com/api/latest/authentication/#validate-api-key
@@ -707,32 +558,16 @@ var (
 	influxWriteRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/influx/write", protocol="influx"}`)
 	influxWriteErrors   = metrics.NewCounter(`vmagent_http_request_errors_total{path="/influx/write", protocol="influx"}`)
 
-	influxQueryRequests  = metrics.NewCounter(`vmagent_http_requests_total{path="/influx/query", protocol="influx"}`)
-	influxHealthRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/influx/health", protocol="influx"}`)
+	influxQueryRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/influx/query", protocol="influx"}`)
 
-	datadogv1WriteRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/datadog/api/v1/series", protocol="datadog"}`)
-	datadogv1WriteErrors   = metrics.NewCounter(`vmagent_http_request_errors_total{path="/datadog/api/v1/series", protocol="datadog"}`)
-
-	datadogv2WriteRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/datadog/api/v2/series", protocol="datadog"}`)
-	datadogv2WriteErrors   = metrics.NewCounter(`vmagent_http_request_errors_total{path="/datadog/api/v2/series", protocol="datadog"}`)
-
-	datadogsketchesWriteRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/datadog/api/beta/sketches", protocol="datadog"}`)
-	datadogsketchesWriteErrors   = metrics.NewCounter(`vmagent_http_request_errors_total{path="/datadog/api/beta/sketches", protocol="datadog"}`)
+	datadogWriteRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/datadog/api/v1/series", protocol="datadog"}`)
+	datadogWriteErrors   = metrics.NewCounter(`vmagent_http_request_errors_total{path="/datadog/api/v1/series", protocol="datadog"}`)
 
 	datadogValidateRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/datadog/api/v1/validate", protocol="datadog"}`)
 	datadogCheckRunRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/datadog/api/v1/check_run", protocol="datadog"}`)
 	datadogIntakeRequests   = metrics.NewCounter(`vmagent_http_requests_total{path="/datadog/intake", protocol="datadog"}`)
 	datadogMetadataRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/datadog/api/v1/metadata", protocol="datadog"}`)
 
-	opentelemetryPushRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/opentelemetry/v1/metrics", protocol="opentelemetry"}`)
-	opentelemetryPushErrors   = metrics.NewCounter(`vmagent_http_request_errors_total{path="/opentelemetry/v1/metrics", protocol="opentelemetry"}`)
-
-	newrelicWriteRequests = metrics.NewCounter(`vm_http_requests_total{path="/newrelic/infra/v2/metrics/events/bulk", protocol="newrelic"}`)
-	newrelicWriteErrors   = metrics.NewCounter(`vm_http_request_errors_total{path="/newrelic/infra/v2/metrics/events/bulk", protocol="newrelic"}`)
-
-	newrelicInventoryRequests = metrics.NewCounter(`vm_http_requests_total{path="/newrelic/inventory/deltas", protocol="newrelic"}`)
-	newrelicCheckRequest      = metrics.NewCounter(`vm_http_requests_total{path="/newrelic", protocol="newrelic"}`)
-
 	promscrapeTargetsRequests          = metrics.NewCounter(`vmagent_http_requests_total{path="/targets"}`)
 	promscrapeServiceDiscoveryRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/service-discovery"}`)
 
@@ -754,7 +589,7 @@ func usage() {
 	const s = `
 vmagent collects metrics data via popular data ingestion protocols and routes it to VictoriaMetrics.
 
-See the docs at https://docs.victoriametrics.com/victoriametrics/vmagent/ .
+See the docs at https://docs.victoriametrics.com/vmagent.html .
 `
 	flagutil.Usage(s)
 }

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=non-existing
-ARG root_image=non-existing
-FROM $certs_image AS certs
+ARG certs_image
+ARG root_image
+FROM $certs_image as certs
 RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
 
 FROM $root_image
@@ -9,5 +9,4 @@ COPY --from=certs /etc/ssl/certs/ca-certificates.crt /etc/ssl/certs/ca-certifica
 EXPOSE 8429
 ENTRYPOINT ["/vmagent-prod"]
 ARG TARGETARCH
-ARG BINARY_SUFFIX=non-existing
-COPY vmagent-linux-${TARGETARCH}-prod${BINARY_SUFFIX} ./vmagent-prod
+COPY vmagent-linux-${TARGETARCH}-prod ./vmagent-prod

app/vmagent/native/request_handler.go

@@ -8,9 +8,9 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/auth"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
+	parserCommon "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/native/stream"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/tenantmetrics"
 	"github.com/VictoriaMetrics/metrics"
 )
@@ -25,17 +25,17 @@ var (
 //
 // See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/6
 func InsertHandler(at *auth.Token, req *http.Request) error {
-	extraLabels, err := protoparserutil.GetExtraLabels(req)
+	extraLabels, err := parserCommon.GetExtraLabels(req)
 	if err != nil {
 		return err
 	}
-	encoding := req.Header.Get("Content-Encoding")
-	return stream.Parse(req.Body, encoding, func(block *stream.Block) error {
+	isGzip := req.Header.Get("Content-Encoding") == "gzip"
+	return stream.Parse(req.Body, isGzip, func(block *stream.Block) error {
 		return insertRows(at, block, extraLabels)
 	})
 }
 
-func insertRows(at *auth.Token, block *stream.Block, extraLabels []prompb.Label) error {
+func insertRows(at *auth.Token, block *stream.Block, extraLabels []prompbmarshal.Label) error {
 	ctx := common.GetPushCtx()
 	defer common.PutPushCtx(ctx)
 
@@ -53,13 +53,13 @@ func insertRows(at *auth.Token, block *stream.Block, extraLabels []prompb.Label)
 	samples := ctx.Samples[:0]
 	mn := &block.MetricName
 	labelsLen := len(labels)
-	labels = append(labels, prompb.Label{
+	labels = append(labels, prompbmarshal.Label{
 		Name:  "__name__",
 		Value: bytesutil.ToUnsafeString(mn.MetricGroup),
 	})
 	for j := range mn.Tags {
 		tag := &mn.Tags[j]
-		labels = append(labels, prompb.Label{
+		labels = append(labels, prompbmarshal.Label{
 			Name:  bytesutil.ToUnsafeString(tag.Key),
 			Value: bytesutil.ToUnsafeString(tag.Value),
 		})
@@ -72,20 +72,18 @@ func insertRows(at *auth.Token, block *stream.Block, extraLabels []prompb.Label)
 	}
 	samplesLen := len(samples)
 	for j, value := range values {
-		samples = append(samples, prompb.Sample{
+		samples = append(samples, prompbmarshal.Sample{
 			Value:     value,
 			Timestamp: timestamps[j],
 		})
 	}
-	tssDst = append(tssDst, prompb.TimeSeries{
+	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
-	}
+	remotewrite.Push(at, &ctx.WriteRequest)
 	return nil
 }

app/vmagent/newrelic/request_handler.go

@@ -1,87 +0,0 @@
-package newrelic
-
-import (
-	"net/http"
-
-	"github.com/VictoriaMetrics/metrics"
-
-	"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/bytesutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/newrelic"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/newrelic/stream"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/tenantmetrics"
-)
-
-var (
-	rowsInserted       = metrics.NewCounter(`vmagent_rows_inserted_total{type="newrelic"}`)
-	rowsTenantInserted = tenantmetrics.NewCounterMap(`vmagent_tenant_inserted_rows_total{type="newrelic"}`)
-	rowsPerInsert      = metrics.NewHistogram(`vmagent_rows_per_insert{type="newrelic"}`)
-)
-
-// InsertHandlerForHTTP processes remote write for NewRelic POST /infra/v2/metrics/events/bulk request.
-func InsertHandlerForHTTP(at *auth.Token, req *http.Request) error {
-	extraLabels, err := protoparserutil.GetExtraLabels(req)
-	if err != nil {
-		return err
-	}
-	encoding := req.Header.Get("Content-Encoding")
-	return stream.Parse(req.Body, encoding, func(rows []newrelic.Row) error {
-		return insertRows(at, rows, extraLabels)
-	})
-}
-
-func insertRows(at *auth.Token, rows []newrelic.Row, extraLabels []prompb.Label) error {
-	ctx := common.GetPushCtx()
-	defer common.PutPushCtx(ctx)
-
-	samplesCount := 0
-	tssDst := ctx.WriteRequest.Timeseries[:0]
-	labels := ctx.Labels[:0]
-	samples := ctx.Samples[:0]
-	for i := range rows {
-		r := &rows[i]
-		tags := r.Tags
-		srcSamples := r.Samples
-		for j := range srcSamples {
-			s := &srcSamples[j]
-			labelsLen := len(labels)
-			labels = append(labels, prompb.Label{
-				Name:  "__name__",
-				Value: bytesutil.ToUnsafeString(s.Name),
-			})
-			for k := range tags {
-				t := &tags[k]
-				labels = append(labels, prompb.Label{
-					Name:  bytesutil.ToUnsafeString(t.Key),
-					Value: bytesutil.ToUnsafeString(t.Value),
-				})
-			}
-			samples = append(samples, prompb.Sample{
-				Value:     s.Value,
-				Timestamp: r.Timestamp,
-			})
-			tssDst = append(tssDst, prompb.TimeSeries{
-				Labels:  labels[labelsLen:],
-				Samples: samples[len(samples)-1:],
-			})
-			labels = append(labels, extraLabels...)
-		}
-		samplesCount += len(srcSamples)
-	}
-	ctx.WriteRequest.Timeseries = tssDst
-	ctx.Labels = labels
-	ctx.Samples = samples
-	if !remotewrite.TryPush(at, &ctx.WriteRequest) {
-		return remotewrite.ErrQueueFullHTTPRetry
-	}
-	rowsInserted.Add(len(rows))
-	if at != nil {
-		rowsTenantInserted.Get(at).Add(samplesCount)
-	}
-	rowsPerInsert.Update(float64(samplesCount))
-	return nil
-}

app/vmagent/opentelemetry/request_handler.go

@@ -1,99 +0,0 @@
-package opentelemetry
-
-import (
-	"fmt"
-	"net/http"
-
-	"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/prompb"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promscrape"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentelemetry/firehose"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentelemetry/stream"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/tenantmetrics"
-	"github.com/VictoriaMetrics/metrics"
-)
-
-var (
-	rowsInserted           = metrics.NewCounter(`vmagent_rows_inserted_total{type="opentelemetry"}`)
-	metadataInserted       = metrics.NewCounter(`vmagent_metadata_inserted_total{type="opentelemetry"}`)
-	rowsTenantInserted     = tenantmetrics.NewCounterMap(`vmagent_tenant_inserted_rows_total{type="opentelemetry"}`)
-	metadataTenantInserted = tenantmetrics.NewCounterMap(`vmagent_tenant_inserted_metadata_total{type="opentelemetry"}`)
-	rowsPerInsert          = metrics.NewHistogram(`vmagent_rows_per_insert{type="opentelemetry"}`)
-)
-
-// InsertHandler processes opentelemetry metrics.
-func InsertHandler(at *auth.Token, req *http.Request) error {
-	extraLabels, err := protoparserutil.GetExtraLabels(req)
-	if err != nil {
-		return err
-	}
-	encoding := req.Header.Get("Content-Encoding")
-	var processBody func([]byte) ([]byte, error)
-	if req.Header.Get("Content-Type") == "application/json" {
-		if req.Header.Get("X-Amz-Firehose-Protocol-Version") != "" {
-			processBody = firehose.ProcessRequestBody
-		} else {
-			return fmt.Errorf("json encoding isn't supported for opentelemetry format. Use protobuf encoding")
-		}
-	}
-	return stream.ParseStream(req.Body, encoding, processBody, func(tss []prompb.TimeSeries, mms []prompb.MetricMetadata) error {
-		return insertRows(at, tss, mms, extraLabels)
-	})
-}
-
-func insertRows(at *auth.Token, tss []prompb.TimeSeries, mms []prompb.MetricMetadata, extraLabels []prompb.Label) error {
-	ctx := common.GetPushCtx()
-	defer common.PutPushCtx(ctx)
-
-	rowsTotal := 0
-	tssDst := ctx.WriteRequest.Timeseries[:0]
-	labels := ctx.Labels[:0]
-	samples := ctx.Samples[:0]
-	for i := range tss {
-		ts := &tss[i]
-		rowsTotal += len(ts.Samples)
-		labelsLen := len(labels)
-		labels = append(labels, ts.Labels...)
-		labels = append(labels, extraLabels...)
-		samplesLen := len(samples)
-		samples = append(samples, ts.Samples...)
-		tssDst = append(tssDst, prompb.TimeSeries{
-			Labels:  labels[labelsLen:],
-			Samples: samples[samplesLen:],
-		})
-	}
-	ctx.WriteRequest.Timeseries = tssDst
-
-	var metadataTotal int
-	if promscrape.IsMetadataEnabled() {
-		var accountID, projectID uint32
-		if at != nil {
-			accountID = at.AccountID
-			projectID = at.ProjectID
-			for i := range mms {
-				mm := &mms[i]
-				mm.AccountID = accountID
-				mm.ProjectID = projectID
-			}
-		}
-		ctx.WriteRequest.Metadata = mms
-		metadataTotal = len(mms)
-	}
-
-	ctx.Labels = labels
-	ctx.Samples = samples
-	if !remotewrite.TryPush(at, &ctx.WriteRequest) {
-		return remotewrite.ErrQueueFullHTTPRetry
-	}
-	rowsInserted.Add(rowsTotal)
-	metadataInserted.Add(metadataTotal)
-	if at != nil {
-		rowsTenantInserted.Get(at).Add(rowsTotal)
-		metadataTenantInserted.Get(at).Add(metadataTotal)
-	}
-	rowsPerInsert.Update(float64(rowsTotal))
-	return nil
-}

app/vmagent/opentsdb/request_handler.go

@@ -5,7 +5,7 @@ import (
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/common"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/remotewrite"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
 	parser "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentsdb"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentsdb/stream"
 	"github.com/VictoriaMetrics/metrics"
@@ -33,22 +33,22 @@ func insertRows(rows []parser.Row) error {
 	for i := range rows {
 		r := &rows[i]
 		labelsLen := len(labels)
-		labels = append(labels, prompb.Label{
+		labels = append(labels, prompbmarshal.Label{
 			Name:  "__name__",
 			Value: r.Metric,
 		})
 		for j := range r.Tags {
 			tag := &r.Tags[j]
-			labels = append(labels, prompb.Label{
+			labels = append(labels, prompbmarshal.Label{
 				Name:  tag.Key,
 				Value: tag.Value,
 			})
 		}
-		samples = append(samples, prompb.Sample{
+		samples = append(samples, prompbmarshal.Sample{
 			Value:     r.Value,
 			Timestamp: r.Timestamp,
 		})
-		tssDst = append(tssDst, prompb.TimeSeries{
+		tssDst = append(tssDst, prompbmarshal.TimeSeries{
 			Labels:  labels[labelsLen:],
 			Samples: samples[len(samples)-1:],
 		})
@@ -56,9 +56,7 @@ func insertRows(rows []parser.Row) error {
 	ctx.WriteRequest.Timeseries = tssDst
 	ctx.Labels = labels
 	ctx.Samples = samples
-	if !remotewrite.TryPush(nil, &ctx.WriteRequest) {
-		return remotewrite.ErrQueueFullHTTPRetry
-	}
+	remotewrite.Push(nil, &ctx.WriteRequest)
 	rowsInserted.Add(len(rows))
 	rowsPerInsert.Update(float64(len(rows)))
 	return nil

app/vmagent/opentsdbhttp/request_handler.go

@@ -6,10 +6,10 @@ import (
 	"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/prompb"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentsdbhttp"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
+	parserCommon "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
+	parser "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentsdbhttp"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentsdbhttp/stream"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
 	"github.com/VictoriaMetrics/metrics"
 )
 
@@ -21,16 +21,16 @@ var (
 // InsertHandler processes HTTP OpenTSDB put requests.
 // See http://opentsdb.net/docs/build/html/api_http/put.html
 func InsertHandler(at *auth.Token, req *http.Request) error {
-	extraLabels, err := protoparserutil.GetExtraLabels(req)
+	extraLabels, err := parserCommon.GetExtraLabels(req)
 	if err != nil {
 		return err
 	}
-	return stream.Parse(req, func(rows []opentsdbhttp.Row) error {
+	return stream.Parse(req, func(rows []parser.Row) error {
 		return insertRows(at, rows, extraLabels)
 	})
 }
 
-func insertRows(at *auth.Token, rows []opentsdbhttp.Row, extraLabels []prompb.Label) error {
+func insertRows(at *auth.Token, rows []parser.Row, extraLabels []prompbmarshal.Label) error {
 	ctx := common.GetPushCtx()
 	defer common.PutPushCtx(ctx)
 
@@ -40,23 +40,23 @@ func insertRows(at *auth.Token, rows []opentsdbhttp.Row, extraLabels []prompb.La
 	for i := range rows {
 		r := &rows[i]
 		labelsLen := len(labels)
-		labels = append(labels, prompb.Label{
+		labels = append(labels, prompbmarshal.Label{
 			Name:  "__name__",
 			Value: r.Metric,
 		})
 		for j := range r.Tags {
 			tag := &r.Tags[j]
-			labels = append(labels, prompb.Label{
+			labels = append(labels, prompbmarshal.Label{
 				Name:  tag.Key,
 				Value: tag.Value,
 			})
 		}
 		labels = append(labels, extraLabels...)
-		samples = append(samples, prompb.Sample{
+		samples = append(samples, prompbmarshal.Sample{
 			Value:     r.Value,
 			Timestamp: r.Timestamp,
 		})
-		tssDst = append(tssDst, prompb.TimeSeries{
+		tssDst = append(tssDst, prompbmarshal.TimeSeries{
 			Labels:  labels[labelsLen:],
 			Samples: samples[len(samples)-1:],
 		})
@@ -64,9 +64,7 @@ func insertRows(at *auth.Token, rows []opentsdbhttp.Row, extraLabels []prompb.La
 	ctx.WriteRequest.Timeseries = tssDst
 	ctx.Labels = labels
 	ctx.Samples = samples
-	if !remotewrite.TryPush(at, &ctx.WriteRequest) {
-		return remotewrite.ErrQueueFullHTTPRetry
-	}
+	remotewrite.Push(at, &ctx.WriteRequest)
 	rowsInserted.Add(len(rows))
 	rowsPerInsert.Update(float64(len(rows)))
 	return nil

app/vmagent/prometheusimport/request_handler.go

@@ -7,103 +7,76 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/remotewrite"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/auth"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promscrape"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/prometheus"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
+	parserCommon "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
+	parser "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/prometheus"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/prometheus/stream"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/tenantmetrics"
 	"github.com/VictoriaMetrics/metrics"
 )
 
 var (
-	rowsInserted           = metrics.NewCounter(`vmagent_rows_inserted_total{type="prometheus"}`)
-	metadataInserted       = metrics.NewCounter(`vmagent_metadata_inserted_total{type="prometheus"}`)
-	rowsTenantInserted     = tenantmetrics.NewCounterMap(`vmagent_tenant_inserted_rows_total{type="prometheus"}`)
-	metadataTenantInserted = tenantmetrics.NewCounterMap(`vmagent_tenant_inserted_metadata_total{type="prometheus"}`)
-
-	rowsPerInsert = metrics.NewHistogram(`vmagent_rows_per_insert{type="prometheus"}`)
+	rowsInserted       = metrics.NewCounter(`vmagent_rows_inserted_total{type="prometheus"}`)
+	rowsTenantInserted = tenantmetrics.NewCounterMap(`vmagent_tenant_inserted_rows_total{type="prometheus"}`)
+	rowsPerInsert      = metrics.NewHistogram(`vmagent_rows_per_insert{type="prometheus"}`)
 )
 
 // InsertHandler processes `/api/v1/import/prometheus` request.
 func InsertHandler(at *auth.Token, req *http.Request) error {
-	extraLabels, err := protoparserutil.GetExtraLabels(req)
+	extraLabels, err := parserCommon.GetExtraLabels(req)
 	if err != nil {
 		return err
 	}
-	defaultTimestamp, err := protoparserutil.GetTimestamp(req)
+	defaultTimestamp, err := parserCommon.GetTimestamp(req)
 	if err != nil {
 		return err
 	}
-	encoding := req.Header.Get("Content-Encoding")
-	return stream.Parse(req.Body, defaultTimestamp, encoding, true, promscrape.IsMetadataEnabled(), func(rows []prometheus.Row, mms []prometheus.Metadata) error {
-		return insertRows(at, rows, mms, extraLabels)
+	isGzipped := req.Header.Get("Content-Encoding") == "gzip"
+	return stream.Parse(req.Body, defaultTimestamp, isGzipped, func(rows []parser.Row) error {
+		return insertRows(at, rows, extraLabels)
 	}, func(s string) {
 		httpserver.LogError(req, s)
 	})
 }
 
-func insertRows(at *auth.Token, rows []prometheus.Row, mms []prometheus.Metadata, extraLabels []prompb.Label) error {
+func insertRows(at *auth.Token, rows []parser.Row, extraLabels []prompbmarshal.Label) error {
 	ctx := common.GetPushCtx()
 	defer common.PutPushCtx(ctx)
 
 	tssDst := ctx.WriteRequest.Timeseries[:0]
-	mmsDst := ctx.WriteRequest.Metadata[:0]
 	labels := ctx.Labels[:0]
 	samples := ctx.Samples[:0]
 	for i := range rows {
 		r := &rows[i]
 		labelsLen := len(labels)
-		labels = append(labels, prompb.Label{
+		labels = append(labels, prompbmarshal.Label{
 			Name:  "__name__",
 			Value: r.Metric,
 		})
 		for j := range r.Tags {
 			tag := &r.Tags[j]
-			labels = append(labels, prompb.Label{
+			labels = append(labels, prompbmarshal.Label{
 				Name:  tag.Key,
 				Value: tag.Value,
 			})
 		}
 		labels = append(labels, extraLabels...)
-		samples = append(samples, prompb.Sample{
+		samples = append(samples, prompbmarshal.Sample{
 			Value:     r.Value,
 			Timestamp: r.Timestamp,
 		})
-		tssDst = append(tssDst, prompb.TimeSeries{
+		tssDst = append(tssDst, prompbmarshal.TimeSeries{
 			Labels:  labels[labelsLen:],
 			Samples: samples[len(samples)-1:],
 		})
 	}
-	var accountID, projectID uint32
-	if at != nil {
-		accountID = at.AccountID
-		projectID = at.ProjectID
-	}
-	for i := range mms {
-		mm := &mms[i]
-		mmsDst = append(mmsDst, prompb.MetricMetadata{
-			MetricFamilyName: mm.Metric,
-			Help:             mm.Help,
-			Type:             mm.Type,
-			// there is no unit in Prometheus exposition formats
-
-			AccountID: accountID,
-			ProjectID: projectID,
-		})
-	}
 	ctx.WriteRequest.Timeseries = tssDst
-	ctx.WriteRequest.Metadata = mmsDst
 	ctx.Labels = labels
 	ctx.Samples = samples
-	if !remotewrite.TryPush(at, &ctx.WriteRequest) {
-		return remotewrite.ErrQueueFullHTTPRetry
-	}
+	remotewrite.Push(at, &ctx.WriteRequest)
 	rowsInserted.Add(len(rows))
-	metadataInserted.Add(len(mms))
 	if at != nil {
 		rowsTenantInserted.Get(at).Add(len(rows))
-		metadataTenantInserted.Get(at).Add(len(mms))
 	}
 	rowsPerInsert.Update(float64(len(rows)))
 	return nil

app/vmagent/prometheusimport/request_handler_test.go

@@ -12,7 +12,7 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/remotewrite"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/fs"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
 )
 
 var (
@@ -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.Fatalf("unexpected error %s", err)
+		t.Errorf("unxepected error %s", err)
 	}
 	expectedMsg := "cannot unmarshal Prometheus line"
 	if !strings.Contains(testOutput.String(), expectedMsg) {
-		t.Fatalf("output %q should contain %q", testOutput.String(), expectedMsg)
+		t.Errorf("output %q should contain %q", testOutput.String(), expectedMsg)
 	}
 }
 
@@ -44,17 +44,17 @@ func setUp() {
 		log.Fatalf("unable to set %q with value %q, err: %v", remoteWriteFlag, srv.URL, err)
 	}
 	logger.Init()
-	protoparserutil.StartUnmarshalWorkers()
+	common.StartUnmarshalWorkers()
 	remotewrite.Init()
 	testOutput = &bytes.Buffer{}
 	logger.SetOutputForTests(testOutput)
 }
 
 func tearDown() {
-	protoparserutil.StopUnmarshalWorkers()
+	common.StopUnmarshalWorkers()
 	srv.Close()
 	logger.ResetOutputForTest()
 	tmpDataDir := flag.Lookup("remoteWrite.tmpDataPath").Value.String()
-	fs.MustRemoveDir(tmpDataDir)
+	fs.MustRemoveAll(tmpDataDir)
 
 }

app/vmagent/promremotewrite/request_handler.go

@@ -6,41 +6,39 @@ import (
 	"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/bytesutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promscrape"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
+	parserCommon "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/promremotewrite/stream"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/tenantmetrics"
 	"github.com/VictoriaMetrics/metrics"
 )
 
 var (
-	rowsInserted           = metrics.NewCounter(`vmagent_rows_inserted_total{type="promremotewrite"}`)
-	metadataInserted       = metrics.NewCounter(`vmagent_metadata_inserted_total{type="promremotewrite"}`)
-	rowsTenantInserted     = tenantmetrics.NewCounterMap(`vmagent_tenant_inserted_rows_total{type="promremotewrite"}`)
-	metadataTenantInserted = tenantmetrics.NewCounterMap(`vmagent_tenant_inserted_metadata_total{type="promremotewrite"}`)
-	rowsPerInsert          = metrics.NewHistogram(`vmagent_rows_per_insert{type="promremotewrite"}`)
+	rowsInserted       = metrics.NewCounter(`vmagent_rows_inserted_total{type="promremotewrite"}`)
+	rowsTenantInserted = tenantmetrics.NewCounterMap(`vmagent_tenant_inserted_rows_total{type="promremotewrite"}`)
+	rowsPerInsert      = metrics.NewHistogram(`vmagent_rows_per_insert{type="promremotewrite"}`)
 )
 
 // InsertHandler processes remote write for prometheus.
 func InsertHandler(at *auth.Token, req *http.Request) error {
-	extraLabels, err := protoparserutil.GetExtraLabels(req)
+	extraLabels, err := parserCommon.GetExtraLabels(req)
 	if err != nil {
 		return err
 	}
 	isVMRemoteWrite := req.Header.Get("Content-Encoding") == "zstd"
-	return stream.Parse(req.Body, isVMRemoteWrite, func(tss []prompb.TimeSeries, mms []prompb.MetricMetadata) error {
-		return insertRows(at, tss, mms, extraLabels)
+	return stream.Parse(req.Body, isVMRemoteWrite, func(tss []prompb.TimeSeries) error {
+		return insertRows(at, tss, extraLabels)
 	})
 }
 
-func insertRows(at *auth.Token, timeseries []prompb.TimeSeries, mms []prompb.MetricMetadata, extraLabels []prompb.Label) error {
+func insertRows(at *auth.Token, timeseries []prompb.TimeSeries, extraLabels []prompbmarshal.Label) error {
 	ctx := common.GetPushCtx()
 	defer common.PutPushCtx(ctx)
 
 	rowsTotal := 0
 	tssDst := ctx.WriteRequest.Timeseries[:0]
-	mmsDst := ctx.WriteRequest.Metadata[:0]
 	labels := ctx.Labels[:0]
 	samples := ctx.Samples[:0]
 	for i := range timeseries {
@@ -49,61 +47,33 @@ func insertRows(at *auth.Token, timeseries []prompb.TimeSeries, mms []prompb.Met
 		labelsLen := len(labels)
 		for i := range ts.Labels {
 			label := &ts.Labels[i]
-			labels = append(labels, prompb.Label{
-				Name:  label.Name,
-				Value: label.Value,
+			labels = append(labels, prompbmarshal.Label{
+				Name:  bytesutil.ToUnsafeString(label.Name),
+				Value: bytesutil.ToUnsafeString(label.Value),
 			})
 		}
 		labels = append(labels, extraLabels...)
 		samplesLen := len(samples)
 		for i := range ts.Samples {
 			sample := &ts.Samples[i]
-			samples = append(samples, prompb.Sample{
+			samples = append(samples, prompbmarshal.Sample{
 				Value:     sample.Value,
 				Timestamp: sample.Timestamp,
 			})
 		}
-		tssDst = append(tssDst, prompb.TimeSeries{
+		tssDst = append(tssDst, prompbmarshal.TimeSeries{
 			Labels:  labels[labelsLen:],
 			Samples: samples[samplesLen:],
 		})
 	}
 	ctx.WriteRequest.Timeseries = tssDst
-
-	var metadataTotal int
-	if promscrape.IsMetadataEnabled() {
-		var accountID, projectID uint32
-		if at != nil {
-			accountID = at.AccountID
-			projectID = at.ProjectID
-		}
-		for i := range mms {
-			mm := &mms[i]
-			mmsDst = append(mmsDst, prompb.MetricMetadata{
-				MetricFamilyName: mm.MetricFamilyName,
-				Help:             mm.Help,
-				Type:             mm.Type,
-				Unit:             mm.Unit,
-
-				AccountID: accountID,
-				ProjectID: projectID,
-			})
-		}
-		ctx.WriteRequest.Metadata = mmsDst
-		metadataTotal = len(mms)
-	}
-
 	ctx.Labels = labels
 	ctx.Samples = samples
-	if !remotewrite.TryPush(at, &ctx.WriteRequest) {
-		return remotewrite.ErrQueueFullHTTPRetry
-	}
+	remotewrite.Push(at, &ctx.WriteRequest)
 	rowsInserted.Add(rowsTotal)
 	if at != nil {
 		rowsTenantInserted.Get(at).Add(rowsTotal)
-		metadataTenantInserted.Get(at).Add(metadataTotal)
 	}
-	metadataInserted.Add(metadataTotal)
 	rowsPerInsert.Update(float64(rowsTotal))
 	return nil
 }

app/vmagent/remotewrite/client.go

@@ -2,50 +2,37 @@ package remotewrite
 
 import (
 	"bytes"
-	"errors"
 	"fmt"
 	"io"
 	"net/http"
 	"net/url"
-	"strconv"
 	"strings"
 	"sync"
-	"sync/atomic"
 	"time"
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/awsapi"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/encoding"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/encoding/zstd"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/persistentqueue"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promauth"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/ratelimiter"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/timerpool"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/timeutil"
 	"github.com/VictoriaMetrics/metrics"
-	"github.com/golang/snappy"
 )
 
 var (
 	forcePromProto = flagutil.NewArrayBool("remoteWrite.forcePromProto", "Whether to force Prometheus remote write protocol for sending data "+
-		"to the corresponding -remoteWrite.url . See https://docs.victoriametrics.com/victoriametrics/vmagent/#victoriametrics-remote-write-protocol")
+		"to the corresponding -remoteWrite.url . See https://docs.victoriametrics.com/vmagent.html#victoriametrics-remote-write-protocol")
 	forceVMProto = flagutil.NewArrayBool("remoteWrite.forceVMProto", "Whether to force VictoriaMetrics remote write protocol for sending data "+
-		"to the corresponding -remoteWrite.url . See https://docs.victoriametrics.com/victoriametrics/vmagent/#victoriametrics-remote-write-protocol")
+		"to the corresponding -remoteWrite.url . See https://docs.victoriametrics.com/vmagent.html#victoriametrics-remote-write-protocol")
 
-	rateLimit = flagutil.NewArrayInt("remoteWrite.rateLimit", 0, "Optional rate limit in bytes per second for data sent to the corresponding -remoteWrite.url. "+
+	rateLimit = flagutil.NewArrayInt("remoteWrite.rateLimit", "Optional rate limit in bytes per second for data sent to the corresponding -remoteWrite.url. "+
 		"By default, the rate limit is disabled. It can be useful for limiting load on remote storage when big amounts of buffered data "+
-		"is sent after temporary unavailability of the remote storage. See also -maxIngestionRate")
-	sendTimeout      = flagutil.NewArrayDuration("remoteWrite.sendTimeout", time.Minute, "Timeout for sending a single block of data to the corresponding -remoteWrite.url")
-	retryMinInterval = flagutil.NewArrayDuration("remoteWrite.retryMinInterval", time.Second, "The minimum delay between retry attempts to send a block of data to the corresponding -remoteWrite.url. Every next retry attempt will double the delay to prevent hammering of remote database. See also -remoteWrite.retryMaxInterval")
-	// deprecated in the future. use -remoteWrite.retryMaxInterval instead
-	retryMaxTime     = flagutil.NewArrayDuration("remoteWrite.retryMaxTime", time.Minute, "The max time spent on retry attempts to send a block of data to the corresponding -remoteWrite.url. This flag is deprecated, use -remoteWrite.retryMaxInterval instead")
-	retryMaxInterval = flagutil.NewArrayDuration("remoteWrite.retryMaxInterval", time.Minute, "The maximum delay between retry attempts to send a block of data to the corresponding -remoteWrite.url.  The delay doubles with each retry until this maximum is reached, after which it remains constant. See also -remoteWrite.retryMinInterval")
-	proxyURL         = flagutil.NewArrayString("remoteWrite.proxyURL", "Optional proxy URL for writing data to the corresponding -remoteWrite.url. "+
+		"is sent after temporary unavailability of the remote storage")
+	sendTimeout = flagutil.NewArrayDuration("remoteWrite.sendTimeout", "Timeout for sending a single block of data to the corresponding -remoteWrite.url")
+	proxyURL    = flagutil.NewArrayString("remoteWrite.proxyURL", "Optional proxy URL for writing data to the corresponding -remoteWrite.url. "+
 		"Supported proxies: http, https, socks5. Example: -remoteWrite.proxyURL=socks5://proxy:1234")
 
-	tlsHandshakeTimeout   = flagutil.NewArrayDuration("remoteWrite.tlsHandshakeTimeout", 20*time.Second, "The timeout for establishing tls connections to the corresponding -remoteWrite.url")
 	tlsInsecureSkipVerify = flagutil.NewArrayBool("remoteWrite.tlsInsecureSkipVerify", "Whether to skip tls verification when connecting to the corresponding -remoteWrite.url")
 	tlsCertFile           = flagutil.NewArrayString("remoteWrite.tlsCertFile", "Optional path to client-side TLS certificate file to use when connecting "+
 		"to the corresponding -remoteWrite.url")
@@ -70,10 +57,8 @@ var (
 	oauth2ClientID         = flagutil.NewArrayString("remoteWrite.oauth2.clientID", "Optional OAuth2 clientID to use for the corresponding -remoteWrite.url")
 	oauth2ClientSecret     = flagutil.NewArrayString("remoteWrite.oauth2.clientSecret", "Optional OAuth2 clientSecret to use for the corresponding -remoteWrite.url")
 	oauth2ClientSecretFile = flagutil.NewArrayString("remoteWrite.oauth2.clientSecretFile", "Optional OAuth2 clientSecretFile to use for the corresponding -remoteWrite.url")
-	oauth2EndpointParams   = flagutil.NewArrayString("remoteWrite.oauth2.endpointParams", "Optional OAuth2 endpoint parameters to use for the corresponding -remoteWrite.url . "+
-		`The endpoint parameters must be set in JSON format: {"param1":"value1",...,"paramN":"valueN"}`)
-	oauth2TokenURL = flagutil.NewArrayString("remoteWrite.oauth2.tokenUrl", "Optional OAuth2 tokenURL to use for the corresponding -remoteWrite.url")
-	oauth2Scopes   = flagutil.NewArrayString("remoteWrite.oauth2.scopes", "Optional OAuth2 scopes to use for the corresponding -remoteWrite.url. Scopes must be delimited by ';'")
+	oauth2TokenURL         = flagutil.NewArrayString("remoteWrite.oauth2.tokenUrl", "Optional OAuth2 tokenURL to use for the corresponding -remoteWrite.url")
+	oauth2Scopes           = flagutil.NewArrayString("remoteWrite.oauth2.scopes", "Optional OAuth2 scopes to use for the corresponding -remoteWrite.url. Scopes must be delimited by ';'")
 
 	awsUseSigv4 = flagutil.NewArrayBool("remoteWrite.aws.useSigv4", "Enables SigV4 request signing for the corresponding -remoteWrite.url. "+
 		"It is expected that other -remoteWrite.aws.* command-line flags are set if sigv4 request signing is enabled")
@@ -92,20 +77,16 @@ type client struct {
 	remoteWriteURL string
 
 	// Whether to use VictoriaMetrics remote write protocol for sending the data to remoteWriteURL
-	useVMProto          atomic.Bool
-	canDowngradeVMProto atomic.Bool
+	useVMProto bool
 
 	fq *persistentqueue.FastQueue
 	hc *http.Client
 
-	retryMinInterval time.Duration
-	retryMaxInterval time.Duration
-
 	sendBlock func(block []byte) bool
 	authCfg   *promauth.Config
 	awsCfg    *awsapi.Config
 
-	rl *ratelimiter.RateLimiter
+	rl rateLimiter
 
 	bytesSent       *metrics.Counter
 	blocksSent      *metrics.Counter
@@ -124,20 +105,22 @@ type client struct {
 func newHTTPClient(argIdx int, remoteWriteURL, sanitizedURL string, fq *persistentqueue.FastQueue, concurrency int) *client {
 	authCfg, err := getAuthConfig(argIdx)
 	if err != nil {
-		logger.Fatalf("cannot initialize auth config for -remoteWrite.url=%q: %s", remoteWriteURL, err)
+		logger.Panicf("FATAL: cannot initialize auth config for remoteWrite.url=%q: %s", remoteWriteURL, err)
 	}
+	tlsCfg := authCfg.NewTLSConfig()
 	awsCfg, err := getAWSAPIConfig(argIdx)
 	if err != nil {
-		logger.Fatalf("cannot initialize AWS Config for -remoteWrite.url=%q: %s", remoteWriteURL, err)
+		logger.Fatalf("FATAL: cannot initialize AWS Config for remoteWrite.url=%q: %s", remoteWriteURL, err)
+	}
+	tr := &http.Transport{
+		DialContext:         statDial,
+		TLSClientConfig:     tlsCfg,
+		TLSHandshakeTimeout: 10 * time.Second,
+		MaxConnsPerHost:     2 * concurrency,
+		MaxIdleConnsPerHost: 2 * concurrency,
+		IdleConnTimeout:     time.Minute,
+		WriteBufferSize:     64 * 1024,
 	}
-
-	tr := httputil.NewTransport(false, "vmagent_remotewrite")
-	tr.TLSHandshakeTimeout = tlsHandshakeTimeout.GetOptionalArg(argIdx)
-	tr.MaxConnsPerHost = 2 * concurrency
-	tr.MaxIdleConnsPerHost = 2 * concurrency
-	tr.IdleConnTimeout = time.Minute
-	tr.WriteBufferSize = 64 * 1024
-
 	pURL := proxyURL.GetOptionalArg(argIdx)
 	if len(pURL) > 0 {
 		if !strings.Contains(pURL, "://") {
@@ -150,23 +133,17 @@ func newHTTPClient(argIdx int, remoteWriteURL, sanitizedURL string, fq *persiste
 		tr.Proxy = http.ProxyURL(pu)
 	}
 	hc := &http.Client{
-		Transport: authCfg.NewRoundTripper(tr),
-		Timeout:   sendTimeout.GetOptionalArg(argIdx),
-	}
-	retryMaxIntervalFlag := retryMaxTime
-	if retryMaxInterval.String() != "" {
-		retryMaxIntervalFlag = retryMaxInterval
+		Transport: tr,
+		Timeout:   sendTimeout.GetOptionalArgOrDefault(argIdx, time.Minute),
 	}
 	c := &client{
-		sanitizedURL:     sanitizedURL,
-		remoteWriteURL:   remoteWriteURL,
-		authCfg:          authCfg,
-		awsCfg:           awsCfg,
-		fq:               fq,
-		hc:               hc,
-		retryMinInterval: retryMinInterval.GetOptionalArg(argIdx),
-		retryMaxInterval: retryMaxIntervalFlag.GetOptionalArg(argIdx),
-		stopCh:           make(chan struct{}),
+		sanitizedURL:   sanitizedURL,
+		remoteWriteURL: remoteWriteURL,
+		authCfg:        authCfg,
+		awsCfg:         awsCfg,
+		fq:             fq,
+		hc:             hc,
+		stopCh:         make(chan struct{}),
 	}
 	c.sendBlock = c.sendBlockHTTP
 
@@ -176,25 +153,32 @@ func newHTTPClient(argIdx int, remoteWriteURL, sanitizedURL string, fq *persiste
 		logger.Fatalf("-remoteWrite.useVMProto and -remoteWrite.usePromProto cannot be set simultaneously for -remoteWrite.url=%s", sanitizedURL)
 	}
 	if !useVMProto && !usePromProto {
-		// The VM protocol could be downgraded later at runtime if unsupported media type response status is received.
-		useVMProto = true
-		c.canDowngradeVMProto.Store(true)
+		// Auto-detect whether the remote storage supports VictoriaMetrics remote write protocol.
+		doRequest := func(url string) (*http.Response, error) {
+			return c.doRequest(url, nil)
+		}
+		useVMProto = common.HandleVMProtoClientHandshake(c.remoteWriteURL, doRequest)
+		if !useVMProto {
+			logger.Infof("the remote storage at %q doesn't support VictoriaMetrics remote write protocol. Switching to Prometheus remote write protocol. "+
+				"See https://docs.victoriametrics.com/vmagent.html#victoriametrics-remote-write-protocol", sanitizedURL)
+		}
 	}
-	c.useVMProto.Store(useVMProto)
+	c.useVMProto = useVMProto
 
 	return c
 }
 
 func (c *client) init(argIdx, concurrency int, sanitizedURL string) {
-	limitReached := metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_rate_limit_reached_total{url=%q}`, c.sanitizedURL))
-	if bytesPerSec := rateLimit.GetOptionalArg(argIdx); bytesPerSec > 0 {
+	if bytesPerSec := rateLimit.GetOptionalArgOrDefault(argIdx, 0); bytesPerSec > 0 {
 		logger.Infof("applying %d bytes per second rate limit for -remoteWrite.url=%q", bytesPerSec, sanitizedURL)
-		c.rl = ratelimiter.New(int64(bytesPerSec), limitReached, c.stopCh)
+		c.rl.perSecondLimit = int64(bytesPerSec)
 	}
+	c.rl.limitReached = metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_rate_limit_reached_total{url=%q}`, c.sanitizedURL))
+
 	c.bytesSent = metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_bytes_sent_total{url=%q}`, c.sanitizedURL))
 	c.blocksSent = metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_blocks_sent_total{url=%q}`, c.sanitizedURL))
 	c.rateLimit = metrics.GetOrCreateGauge(fmt.Sprintf(`vmagent_remotewrite_rate_limit{url=%q}`, c.sanitizedURL), func() float64 {
-		return float64(rateLimit.GetOptionalArg(argIdx))
+		return float64(rateLimit.GetOptionalArgOrDefault(argIdx, 0))
 	})
 	c.requestDuration = metrics.GetOrCreateHistogram(fmt.Sprintf(`vmagent_remotewrite_duration_seconds{url=%q}`, c.sanitizedURL))
 	c.requestsOKCount = metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_requests_total{url=%q, status_code="2XX"}`, c.sanitizedURL))
@@ -246,16 +230,10 @@ func getAuthConfig(argIdx int) (*promauth.Config, error) {
 	clientSecret := oauth2ClientSecret.GetOptionalArg(argIdx)
 	clientSecretFile := oauth2ClientSecretFile.GetOptionalArg(argIdx)
 	if clientSecretFile != "" || clientSecret != "" {
-		endpointParamsJSON := oauth2EndpointParams.GetOptionalArg(argIdx)
-		endpointParams, err := flagutil.ParseJSONMap(endpointParamsJSON)
-		if err != nil {
-			return nil, fmt.Errorf("cannot parse JSON for -remoteWrite.oauth2.endpointParams=%s: %w", endpointParamsJSON, err)
-		}
 		oauth2Cfg = &promauth.OAuth2Config{
 			ClientID:         oauth2ClientID.GetOptionalArg(argIdx),
 			ClientSecret:     promauth.NewSecret(clientSecret),
 			ClientSecretFile: clientSecretFile,
-			EndpointParams:   endpointParams,
 			TokenURL:         oauth2TokenURL.GetOptionalArg(argIdx),
 			Scopes:           strings.Split(oauth2Scopes.GetOptionalArg(argIdx), ";"),
 		}
@@ -279,7 +257,7 @@ func getAuthConfig(argIdx int) (*promauth.Config, error) {
 	}
 	authCfg, err := opts.NewConfig()
 	if err != nil {
-		return nil, fmt.Errorf("cannot populate auth config for remoteWrite idx: %d, err: %w", argIdx, err)
+		return nil, fmt.Errorf("cannot populate OAuth2 config for remoteWrite idx: %d, err: %w", argIdx, err)
 	}
 	return authCfg, nil
 }
@@ -311,11 +289,6 @@ func (c *client) runWorker() {
 		if !ok {
 			return
 		}
-		if len(block) == 0 {
-			// skip empty data blocks from sending
-			// see https://github.com/VictoriaMetrics/VictoriaMetrics/pull/6241
-			continue
-		}
 		go func() {
 			startTime := time.Now()
 			ch <- c.sendBlock(block)
@@ -328,7 +301,7 @@ func (c *client) runWorker() {
 				continue
 			}
 			// Return unsent block to the queue.
-			c.fq.MustWriteBlockIgnoreDisabledPQ(block)
+			c.fq.MustWriteBlock(block)
 			return
 		case <-c.stopCh:
 			// c must be stopped. Wait for a while in the hope the block will be sent.
@@ -337,11 +310,11 @@ func (c *client) runWorker() {
 			case ok := <-ch:
 				if !ok {
 					// Return unsent block to the queue.
-					c.fq.MustWriteBlockIgnoreDisabledPQ(block)
+					c.fq.MustWriteBlock(block)
 				}
 			case <-time.After(graceDuration):
 				// Return unsent block to the queue.
-				c.fq.MustWriteBlockIgnoreDisabledPQ(block)
+				c.fq.MustWriteBlock(block)
 			}
 			return
 		}
@@ -349,46 +322,16 @@ func (c *client) runWorker() {
 }
 
 func (c *client) doRequest(url string, body []byte) (*http.Response, error) {
-	req, err := c.newRequest(url, body)
-	if err != nil {
-		return nil, err
-	}
-	resp, err := c.hc.Do(req)
-	if err == nil {
-		return resp, nil
-	}
-	if !errors.Is(err, io.EOF) && !errors.Is(err, io.ErrUnexpectedEOF) {
-		return nil, err
-	}
-	// It is likely connection became stale or timed out during the first request.
-	// Make another attempt in hope request will succeed.
-	// If not, the error should be handled by the caller as usual.
-	// This should help with https://github.com/VictoriaMetrics/VictoriaMetrics/issues/4139
-	req, err = c.newRequest(url, body)
-	if err != nil {
-		return nil, fmt.Errorf("second attempt: %w", err)
-	}
-	resp, err = c.hc.Do(req)
-	if err != nil {
-		return nil, fmt.Errorf("second attempt: %w", err)
-	}
-	return resp, nil
-}
-
-func (c *client) newRequest(url string, body []byte) (*http.Request, error) {
 	reqBody := bytes.NewBuffer(body)
 	req, err := http.NewRequest(http.MethodPost, url, reqBody)
 	if err != nil {
 		logger.Panicf("BUG: unexpected error from http.NewRequest(%q): %s", url, err)
 	}
-	err = c.authCfg.SetHeaders(req, true)
-	if err != nil {
-		return nil, err
-	}
+	c.authCfg.SetHeaders(req, true)
 	h := req.Header
 	h.Set("User-Agent", "vmagent")
 	h.Set("Content-Type", "application/x-protobuf")
-	if encoding.IsZstd(body) {
+	if c.useVMProto {
 		h.Set("Content-Encoding", "zstd")
 		h.Set("X-VictoriaMetrics-Remote-Write-Version", "1")
 	} else {
@@ -398,20 +341,20 @@ func (c *client) newRequest(url string, body []byte) (*http.Request, error) {
 	if c.awsCfg != nil {
 		sigv4Hash := awsapi.HashHex(body)
 		if err := c.awsCfg.SignRequest(req, sigv4Hash); err != nil {
-			return nil, fmt.Errorf("cannot sign remoteWrite request with AWS sigv4: %w", err)
+			// there is no need in retry, request will be rejected by client.Do and retried by code below
+			logger.Warnf("cannot sign remoteWrite request with AWS sigv4: %s", err)
 		}
 	}
-	return req, nil
+	return c.hc.Do(req)
 }
 
 // sendBlockHTTP sends the given block to c.remoteWriteURL.
 //
 // The function returns false only if c.stopCh is closed.
-// Otherwise, it tries sending the block to remote storage indefinitely.
+// Otherwise it tries sending the block to remote storage indefinitely.
 func (c *client) sendBlockHTTP(block []byte) bool {
-	c.rl.Register(len(block))
-	maxRetryDuration := timeutil.AddJitterToDuration(c.retryMaxInterval)
-	retryDuration := timeutil.AddJitterToDuration(c.retryMinInterval)
+	c.rl.register(len(block), c.stopCh)
+	retryDuration := time.Second
 	retriesCount := 0
 
 again:
@@ -421,10 +364,10 @@ again:
 	if err != nil {
 		c.errorsCount.Inc()
 		retryDuration *= 2
-		if retryDuration > maxRetryDuration {
-			retryDuration = maxRetryDuration
+		if retryDuration > time.Minute {
+			retryDuration = time.Minute
 		}
-		remoteWriteRetryLogger.Warnf("couldn't send a block with size %d bytes to %q: %s; re-sending the block in %.3f seconds",
+		logger.Warnf("couldn't send a block with size %d bytes to %q: %s; re-sending the block in %.3f seconds",
 			len(block), c.sanitizedURL, err, retryDuration.Seconds())
 		t := timerpool.Get(retryDuration)
 		select {
@@ -437,7 +380,6 @@ again:
 		c.retriesCount.Inc()
 		goto again
 	}
-
 	statusCode := resp.StatusCode
 	if statusCode/100 == 2 {
 		_ = resp.Body.Close()
@@ -446,50 +388,21 @@ again:
 		c.blocksSent.Inc()
 		return true
 	}
-
 	metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_requests_total{url=%q, status_code="%d"}`, c.sanitizedURL, statusCode)).Inc()
-	switch statusCode {
-	case 409:
-		logBlockRejected(block, c.sanitizedURL, resp)
-
-		// Just drop block on 409 status code like Prometheus does.
-		// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/873
-		// and https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1149
+	if statusCode == 409 || statusCode == 400 {
+		body, err := io.ReadAll(resp.Body)
 		_ = resp.Body.Close()
-		c.packetsDropped.Inc()
-		return true
-		// - Remote Write v1 specification implicitly expects a `400 Bad Request` when the encoding is not supported.
-		// - Remote Write v2 specification explicitly specifies a `415 Unsupported Media Type` for unsupported encodings.
-		// - Real-world implementations of v1 use both 400 and 415 status codes.
-		// See more in research: https://github.com/VictoriaMetrics/VictoriaMetrics/pull/8462#issuecomment-2786918054
-	case 415, 400:
-		if encoding.IsZstd(block) {
-			logger.Infof("received unsupported media type or bad request from remote storage at %q. Re-packing the block to Prometheus remote write and retrying."+
-				"See https://docs.victoriametrics.com/victoriametrics/vmagent/#victoriametrics-remote-write-protocol", c.sanitizedURL)
-
-			zstdBlockLen := len(block)
-			block, err = repackBlockFromZstdToSnappy(block)
-			if err == nil {
-				if c.canDowngradeVMProto.Swap(false) {
-					logger.Infof("received unsupported media type or bad request from remote storage at %q. Downgrading protocol from VictoriaMetrics to Prometheus remote write for all future requests. "+
-						"See https://docs.victoriametrics.com/victoriametrics/vmagent/#victoriametrics-remote-write-protocol", c.sanitizedURL)
-					c.useVMProto.Store(false)
-				}
-
-				c.retriesCount.Inc()
-				_ = resp.Body.Close()
-				goto again
-			}
-
-			logger.Warnf("failed to repack zstd block (%s bytes) to snappy: %s; The block will be rejected. "+
-				"Possible cause: ungraceful shutdown leading to persisted queue corruption.",
-				zstdBlockLen, err)
+		if err != nil {
+			remoteWriteRejectedLogger.Errorf("sending a block with size %d bytes to %q was rejected (skipping the block): status code %d; "+
+				"failed to read response body: %s",
+				len(block), c.sanitizedURL, statusCode, err)
+		} else {
+			remoteWriteRejectedLogger.Errorf("sending a block with size %d bytes to %q was rejected (skipping the block): status code %d; response body: %s",
+				len(block), c.sanitizedURL, statusCode, string(body))
 		}
-
-		// Just drop snappy blocks on 400 or 415 status codes like Prometheus does.
+		// Just drop block on 409 and 400 status codes like Prometheus does.
 		// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/873
 		// and https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1149
-		logBlockRejected(block, c.sanitizedURL, resp)
 		_ = resp.Body.Close()
 		c.packetsDropped.Inc()
 		return true
@@ -497,10 +410,10 @@ again:
 
 	// Unexpected status code returned
 	retriesCount++
-	retryAfterHeader := parseRetryAfterHeader(resp.Header.Get("Retry-After"))
-	retryDuration = getRetryDuration(retryAfterHeader, retryDuration, maxRetryDuration)
-
-	// Handle response
+	retryDuration *= 2
+	if retryDuration > time.Minute {
+		retryDuration = time.Minute
+	}
 	body, err := io.ReadAll(resp.Body)
 	_ = resp.Body.Close()
 	if err != nil {
@@ -522,79 +435,45 @@ again:
 }
 
 var remoteWriteRejectedLogger = logger.WithThrottler("remoteWriteRejected", 5*time.Second)
-var remoteWriteRetryLogger = logger.WithThrottler("remoteWriteRetry", 5*time.Second)
 
-// getRetryDuration returns retry duration.
-// retryAfterDuration has the highest priority.
-// If retryAfterDuration is not specified, retryDuration gets doubled.
-// retryDuration can't exceed maxRetryDuration.
-//
-// Also see: https://github.com/VictoriaMetrics/VictoriaMetrics/issues/6097
-func getRetryDuration(retryAfterDuration, retryDuration, maxRetryDuration time.Duration) time.Duration {
-	// retryAfterDuration has the highest priority duration
-	if retryAfterDuration > 0 {
-		return timeutil.AddJitterToDuration(retryAfterDuration)
-	}
+type rateLimiter struct {
+	perSecondLimit int64
 
-	// default backoff retry policy
-	retryDuration *= 2
-	if retryDuration > maxRetryDuration {
-		retryDuration = maxRetryDuration
-	}
+	// mu protects budget and deadline from concurrent access.
+	mu sync.Mutex
 
-	return retryDuration
+	// The current budget. It is increased by perSecondLimit every second.
+	budget int64
+
+	// The next deadline for increasing the budget by perSecondLimit
+	deadline time.Time
+
+	limitReached *metrics.Counter
 }
 
-// repackBlockFromZstdToSnappy repacks the given zstd-compressed block to snappy-compressed block.
-//
-// The input block may be corrupted, for example, if vmagent was shut down ungracefully and
-// failed to properly update the persisted queue files. In such cases, zstd decompression
-// will fail and an error will be returned.
-//
-// For more details, see: https://github.com/VictoriaMetrics/VictoriaMetrics/issues/9417
-func repackBlockFromZstdToSnappy(zstdBlock []byte) ([]byte, error) {
-	plainBlock := make([]byte, 0, len(zstdBlock)*2)
-	plainBlock, err := zstd.Decompress(plainBlock, zstdBlock)
-	if err != nil {
-		return nil, fmt.Errorf("zstd: decompress: %s", err)
+func (rl *rateLimiter) register(dataLen int, stopCh <-chan struct{}) {
+	limit := rl.perSecondLimit
+	if limit <= 0 {
+		return
 	}
 
-	return snappy.Encode(nil, plainBlock), nil
-}
-
-func logBlockRejected(block []byte, sanitizedURL string, resp *http.Response) {
-	body, err := io.ReadAll(resp.Body)
-	if err != nil {
-		remoteWriteRejectedLogger.Errorf("sending a block with size %d bytes to %q was rejected (skipping the block): status code %d; "+
-			"failed to read response body: %s",
-			len(block), sanitizedURL, resp.StatusCode, err)
-	} else {
-		remoteWriteRejectedLogger.Errorf("sending a block with size %d bytes to %q was rejected (skipping the block): status code %d; response body: %s",
-			len(block), sanitizedURL, resp.StatusCode, string(body))
-	}
-}
-
-// parseRetryAfterHeader parses `Retry-After` value retrieved from HTTP response header.
-// retryAfterString should be in either HTTP-date or a number of seconds.
-// It will return time.Duration(0) if `retryAfterString` does not follow RFC 7231.
-func parseRetryAfterHeader(retryAfterString string) (retryAfterDuration time.Duration) {
-	if retryAfterString == "" {
-		return retryAfterDuration
-	}
-
-	defer func() {
-		v := retryAfterDuration.Seconds()
-		logger.Infof("'Retry-After: %s' parsed into %.2f second(s)", retryAfterString, v)
-	}()
-
-	// Retry-After could be in "Mon, 02 Jan 2006 15:04:05 GMT" format.
-	if parsedTime, err := time.Parse(http.TimeFormat, retryAfterString); err == nil {
-		return time.Duration(time.Until(parsedTime).Seconds()) * time.Second
-	}
-	// Retry-After could be in seconds.
-	if seconds, err := strconv.Atoi(retryAfterString); err == nil {
-		return time.Duration(seconds) * time.Second
-	}
-
-	return 0
+	rl.mu.Lock()
+	defer rl.mu.Unlock()
+
+	for rl.budget <= 0 {
+		if d := time.Until(rl.deadline); d > 0 {
+			rl.limitReached.Inc()
+			t := timerpool.Get(d)
+			select {
+			case <-stopCh:
+				timerpool.Put(t)
+				return
+			case <-t.C:
+				timerpool.Put(t)
+			}
+		}
+		rl.budget += limit
+		rl.deadline = time.Now().Add(time.Second)
+	}
+	rl.budget -= int64(dataLen)
 }

app/vmagent/remotewrite/client_test.go

@@ -1,132 +0,0 @@
-package remotewrite
-
-import (
-	"math"
-	"net/http"
-	"testing"
-	"time"
-
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/encoding"
-	"github.com/golang/snappy"
-)
-
-func TestCalculateRetryDuration(t *testing.T) {
-	// `testFunc` call `calculateRetryDuration` for `n` times
-	// and evaluate if the result of `calculateRetryDuration` is
-	// 1. >= expectMinDuration
-	// 2. <= expectMinDuration + 10% (see timeutil.AddJitterToDuration)
-	f := func(retryAfterDuration, retryDuration time.Duration, n int, expectMinDuration time.Duration) {
-		t.Helper()
-
-		for i := 0; i < n; i++ {
-			retryDuration = getRetryDuration(retryAfterDuration, retryDuration, time.Minute)
-		}
-
-		expectMaxDuration := helper(expectMinDuration)
-		expectMinDuration = expectMinDuration - (1000 * time.Millisecond) // Avoid edge case when calculating time.Until(now)
-
-		if retryDuration < expectMinDuration || retryDuration > expectMaxDuration {
-			t.Fatalf(
-				"incorrect retry duration, want (ms): [%d, %d], got (ms): %d",
-				expectMinDuration.Milliseconds(), expectMaxDuration.Milliseconds(),
-				retryDuration.Milliseconds(),
-			)
-		}
-	}
-
-	// Call calculateRetryDuration for 1 time.
-	{
-		// default backoff policy
-		f(0, time.Second, 1, 2*time.Second)
-		// default backoff policy exceed max limit"
-		f(0, 10*time.Minute, 1, time.Minute)
-
-		// retry after > default backoff policy
-		f(10*time.Second, 1*time.Second, 1, 10*time.Second)
-		// retry after < default backoff policy
-		f(1*time.Second, 10*time.Second, 1, 1*time.Second)
-		// retry after invalid and < default backoff policy
-		f(0, time.Second, 1, 2*time.Second)
-
-	}
-
-	// Call calculateRetryDuration for multiple times.
-	{
-		// default backoff policy 2 times
-		f(0, time.Second, 2, 4*time.Second)
-		// default backoff policy 3 times
-		f(0, time.Second, 3, 8*time.Second)
-		// default backoff policy N times exceed max limit
-		f(0, time.Second, 10, time.Minute)
-
-		// retry after 120s 1 times
-		f(120*time.Second, time.Second, 1, 120*time.Second)
-		// retry after 120s 2 times
-		f(120*time.Second, time.Second, 2, 120*time.Second)
-	}
-}
-
-func TestParseRetryAfterHeader(t *testing.T) {
-	f := func(retryAfterString string, expectResult time.Duration) {
-		t.Helper()
-
-		result := parseRetryAfterHeader(retryAfterString)
-		// expect `expectResult == result` when retryAfterString is in seconds or invalid
-		// expect the difference between result and expectResult to be lower than 10%
-		if !(expectResult == result || math.Abs(float64(expectResult-result))/float64(expectResult) < 0.10) {
-			t.Fatalf(
-				"incorrect retry after duration, want (ms): %d, got (ms): %d",
-				expectResult.Milliseconds(), result.Milliseconds(),
-			)
-		}
-	}
-
-	// retry after header in seconds
-	f("10", 10*time.Second)
-	// retry after header in date time
-	f(time.Now().Add(30*time.Second).UTC().Format(http.TimeFormat), 30*time.Second)
-	// retry after header invalid
-	f("invalid-retry-after", 0)
-	// retry after header not in GMT
-	f(time.Now().Add(10*time.Second).Format("Mon, 02 Jan 2006 15:04:05 FAKETZ"), 0)
-}
-
-// helper calculate the max possible time duration calculated by timeutil.AddJitterToDuration.
-func helper(d time.Duration) time.Duration {
-	dv := d / 10
-	if dv > 10*time.Second {
-		dv = 10 * time.Second
-	}
-
-	return d + dv
-}
-
-func TestRepackBlockFromZstdToSnappy(t *testing.T) {
-	expectedPlainBlock := []byte(`foobar`)
-
-	zstdBlock := encoding.CompressZSTDLevel(nil, expectedPlainBlock, 1)
-	snappyBlock, err := repackBlockFromZstdToSnappy(zstdBlock)
-	if err != nil {
-		t.Fatalf("unexpected error: %s", err)
-	}
-
-	actualPlainBlock, err := snappy.Decode(nil, snappyBlock)
-	if err != nil {
-		t.Fatalf("unexpected error: %s", err)
-	}
-
-	if string(actualPlainBlock) != string(expectedPlainBlock) {
-		t.Fatalf("unexpected plain block; got %q; want %q", actualPlainBlock, expectedPlainBlock)
-	}
-}
-
-func TestRepackBlockFromZstdToSnappyInvalidBlock(t *testing.T) {
-	snappyBlock, err := repackBlockFromZstdToSnappy([]byte("invalid zstd block"))
-
-	if err == nil {
-		t.Fatalf("expected error for invalid zstd block; got nil")
-	}
-	if len(snappyBlock) != 0 {
-		t.Fatalf("expected empty snappy block; got %d bytes", len(snappyBlock))
-	}
-}

app/vmagent/remotewrite/pendingseries.go

@@ -7,30 +7,26 @@ import (
 	"time"
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/cgroup"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/decimal"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/encoding/zstd"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/fasttime"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/persistentqueue"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promrelabel"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/slicesutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/timeutil"
 	"github.com/VictoriaMetrics/metrics"
 	"github.com/golang/snappy"
 )
 
 var (
 	flushInterval = flag.Duration("remoteWrite.flushInterval", time.Second, "Interval for flushing the data to remote storage. "+
-		"This option takes effect only when less than -remoteWrite.maxRowsPerBlock data points per -remoteWrite.flushInterval are pushed to -remoteWrite.url")
+		"This option takes effect only when less than 10K data points per second are pushed to -remoteWrite.url")
 	maxUnpackedBlockSize = flagutil.NewBytes("remoteWrite.maxBlockSize", 8*1024*1024, "The maximum block size to send to remote storage. Bigger blocks may improve performance at the cost of the increased memory usage. See also -remoteWrite.maxRowsPerBlock")
 	maxRowsPerBlock      = flag.Int("remoteWrite.maxRowsPerBlock", 10000, "The maximum number of samples to send in each block to remote storage. Higher number may improve performance at the cost of the increased memory usage. See also -remoteWrite.maxBlockSize")
-	maxMetadataPerBlock  = flag.Int("remoteWrite.maxMetadataPerBlock", 5000, "The maximum number of metadata to send in each block to remote storage. Higher number may improve performance at the cost of the increased memory usage. See also -remoteWrite.maxBlockSize")
 	vmProtoCompressLevel = flag.Int("remoteWrite.vmProtoCompressLevel", 0, "The compression level for VictoriaMetrics remote write protocol. "+
 		"Higher values reduce network traffic at the cost of higher CPU usage. Negative values reduce CPU usage at the cost of increased network traffic. "+
-		"See https://docs.victoriametrics.com/victoriametrics/vmagent/#victoriametrics-remote-write-protocol")
+		"See https://docs.victoriametrics.com/vmagent.html#victoriametrics-remote-write-protocol")
 )
 
 type pendingSeries struct {
@@ -41,9 +37,9 @@ type pendingSeries struct {
 	periodicFlusherWG sync.WaitGroup
 }
 
-func newPendingSeries(fq *persistentqueue.FastQueue, isVMRemoteWrite *atomic.Bool, significantFigures, roundDigits int) *pendingSeries {
+func newPendingSeries(pushBlock func(block []byte), isVMRemoteWrite bool, significantFigures, roundDigits int) *pendingSeries {
 	var ps pendingSeries
-	ps.wr.fq = fq
+	ps.wr.pushBlock = pushBlock
 	ps.wr.isVMRemoteWrite = isVMRemoteWrite
 	ps.wr.significantFigures = significantFigures
 	ps.wr.roundDigits = roundDigits
@@ -61,18 +57,10 @@ func (ps *pendingSeries) MustStop() {
 	ps.periodicFlusherWG.Wait()
 }
 
-func (ps *pendingSeries) TryPushTimeSeries(tss []prompb.TimeSeries) bool {
+func (ps *pendingSeries) Push(tss []prompbmarshal.TimeSeries) {
 	ps.mu.Lock()
-	ok := ps.wr.tryPushTimeSeries(tss)
+	ps.wr.push(tss)
 	ps.mu.Unlock()
-	return ok
-}
-
-func (ps *pendingSeries) TryPushMetadata(mms []prompb.MetricMetadata) bool {
-	ps.mu.Lock()
-	ok := ps.wr.tryPushMetadata(mms)
-	ps.mu.Unlock()
-	return ok
 }
 
 func (ps *pendingSeries) periodicFlusher() {
@@ -80,111 +68,85 @@ func (ps *pendingSeries) periodicFlusher() {
 	if flushSeconds <= 0 {
 		flushSeconds = 1
 	}
-	d := timeutil.AddJitterToDuration(*flushInterval)
-	ticker := time.NewTicker(d)
+	ticker := time.NewTicker(*flushInterval)
 	defer ticker.Stop()
-	for {
+	mustStop := false
+	for !mustStop {
 		select {
 		case <-ps.stopCh:
-			ps.mu.Lock()
-			ps.wr.mustFlushOnStop()
-			ps.mu.Unlock()
-			return
+			mustStop = true
 		case <-ticker.C:
-			if fasttime.UnixTimestamp()-ps.wr.lastFlushTime.Load() < uint64(flushSeconds) {
+			if fasttime.UnixTimestamp()-atomic.LoadUint64(&ps.wr.lastFlushTime) < uint64(flushSeconds) {
 				continue
 			}
 		}
 		ps.mu.Lock()
-		_ = ps.wr.tryFlush()
+		ps.wr.flush()
 		ps.mu.Unlock()
 	}
 }
 
 type writeRequest struct {
-	lastFlushTime atomic.Uint64
+	// Move lastFlushTime to the top of the struct in order to guarantee atomic access on 32-bit architectures.
+	lastFlushTime uint64
 
-	// The queue to send blocks to.
-	fq *persistentqueue.FastQueue
+	// pushBlock is called when whe write request is ready to be sent.
+	pushBlock func(block []byte)
 
 	// Whether to encode the write request with VictoriaMetrics remote write protocol.
-	isVMRemoteWrite *atomic.Bool
+	isVMRemoteWrite bool
 
-	// How many significant figures must be left before sending the writeRequest to fq.
+	// How many significant figures must be left before sending the writeRequest to pushBlock.
 	significantFigures int
 
-	// How many decimal digits after point must be left before sending the writeRequest to fq.
+	// How many decimal digits after point must be left before sending the writeRequest to pushBlock.
 	roundDigits int
 
-	wr prompb.WriteRequest
+	wr prompbmarshal.WriteRequest
 
-	tss     []prompb.TimeSeries
-	mms     []prompb.MetricMetadata
-	labels  []prompb.Label
-	samples []prompb.Sample
+	tss []prompbmarshal.TimeSeries
 
-	// buf holds labels data
-	buf []byte
-	// metadatabuf holds metadata data
-	metadatabuf []byte
+	labels  []prompbmarshal.Label
+	samples []prompbmarshal.Sample
+	buf     []byte
 }
 
 func (wr *writeRequest) reset() {
-	// Do not reset lastFlushTime, fq, isVMRemoteWrite, significantFigures and roundDigits, since they are reused.
+	// Do not reset lastFlushTime, pushBlock, isVMRemoteWrite, significantFigures and roundDigits, since they are re-used.
 
 	wr.wr.Timeseries = nil
-	wr.wr.Metadata = nil
 
-	clear(wr.tss)
+	for i := range wr.tss {
+		ts := &wr.tss[i]
+		ts.Labels = nil
+		ts.Samples = nil
+	}
 	wr.tss = wr.tss[:0]
 
-	clear(wr.mms)
-	wr.mms = wr.mms[:0]
-
 	promrelabel.CleanLabels(wr.labels)
 	wr.labels = wr.labels[:0]
 
 	wr.samples = wr.samples[:0]
 	wr.buf = wr.buf[:0]
-	wr.metadatabuf = wr.metadatabuf[:0]
 }
 
-// mustFlushOnStop force pushes wr data into wr.fq
-//
-// This is needed in order to properly save in-memory data to persistent queue on graceful shutdown.
-func (wr *writeRequest) mustFlushOnStop() {
+func (wr *writeRequest) flush() {
 	wr.wr.Timeseries = wr.tss
-	wr.wr.Metadata = wr.mms
-	if !tryPushWriteRequest(&wr.wr, wr.mustWriteBlock, wr.isVMRemoteWrite.Load()) {
-		logger.Panicf("BUG: final flush must always return true")
-	}
+	wr.adjustSampleValues()
+	atomic.StoreUint64(&wr.lastFlushTime, fasttime.UnixTimestamp())
+	pushWriteRequest(&wr.wr, wr.pushBlock, wr.isVMRemoteWrite)
 	wr.reset()
 }
 
-func (wr *writeRequest) mustWriteBlock(block []byte) bool {
-	wr.fq.MustWriteBlockIgnoreDisabledPQ(block)
-	return true
-}
-
-func (wr *writeRequest) tryFlush() bool {
-	wr.wr.Timeseries = wr.tss
-	wr.wr.Metadata = wr.mms
-	wr.lastFlushTime.Store(fasttime.UnixTimestamp())
-	if !tryPushWriteRequest(&wr.wr, wr.fq.TryWriteBlock, wr.isVMRemoteWrite.Load()) {
-		return false
-	}
-	wr.reset()
-	return true
-}
-
-func adjustSampleValues(samples []prompb.Sample, significantFigures, roundDigits int) {
-	if n := significantFigures; n > 0 {
+func (wr *writeRequest) adjustSampleValues() {
+	samples := wr.samples
+	if n := wr.significantFigures; n > 0 {
 		for i := range samples {
 			s := &samples[i]
 			s.Value = decimal.RoundToSignificantFigures(s.Value, n)
 		}
 	}
-	if n := roundDigits; n < 100 {
+	if n := wr.roundDigits; n < 100 {
 		for i := range samples {
 			s := &samples[i]
 			s.Value = decimal.RoundToDecimalDigits(s.Value, n)
@@ -192,236 +154,104 @@ func adjustSampleValues(samples []prompb.Sample, significantFigures, roundDigits
 	}
 }
 
-func (wr *writeRequest) tryPushMetadata(mms []prompb.MetricMetadata) bool {
-	mmdDst := wr.mms
-	maxMetadataPerBlock := *maxMetadataPerBlock
-	for i := range mms {
-		if len(wr.mms) >= maxMetadataPerBlock {
-			if !wr.tryFlush() {
-				return false
-			}
-			mmdDst = wr.mms
-		}
-		mmSrc := &mms[i]
-		mmdDst = append(mmdDst, prompb.MetricMetadata{})
-		wr.copyMetadata(&mmdDst[len(mmdDst)-1], mmSrc)
-	}
-	wr.mms = mmdDst
-	return true
-}
-
-func (wr *writeRequest) copyMetadata(dst, src *prompb.MetricMetadata) {
-	// Direct copy for non-string fields, which are safe by value.
-	dst.Type = src.Type
-	dst.Unit = src.Unit
-
-	// Pre-allocate memory for all string fields.
-	neededBufLen := len(src.MetricFamilyName) + len(src.Help)
-	bufLen := len(wr.metadatabuf)
-	wr.metadatabuf = slicesutil.SetLength(wr.metadatabuf, bufLen+neededBufLen)
-	buf := wr.metadatabuf[:bufLen]
-
-	// Copy MetricFamilyName
-	bufLen = len(buf)
-	buf = append(buf, src.MetricFamilyName...)
-	dst.MetricFamilyName = bytesutil.ToUnsafeString(buf[bufLen:])
-
-	// Copy Help
-	bufLen = len(buf)
-	buf = append(buf, src.Help...)
-	dst.Help = bytesutil.ToUnsafeString(buf[bufLen:])
-
-	wr.metadatabuf = buf
-}
-
-func (wr *writeRequest) tryPushTimeSeries(src []prompb.TimeSeries) bool {
+func (wr *writeRequest) push(src []prompbmarshal.TimeSeries) {
 	tssDst := wr.tss
 	maxSamplesPerBlock := *maxRowsPerBlock
 	// Allow up to 10x of labels per each block on average.
 	maxLabelsPerBlock := 10 * maxSamplesPerBlock
 	for i := range src {
+		tssDst = append(tssDst, prompbmarshal.TimeSeries{})
+		wr.copyTimeSeries(&tssDst[len(tssDst)-1], &src[i])
 		if len(wr.samples) >= maxSamplesPerBlock || len(wr.labels) >= maxLabelsPerBlock {
 			wr.tss = tssDst
-			if !wr.tryFlush() {
-				return false
-			}
+			wr.flush()
 			tssDst = wr.tss
 		}
-		tsSrc := &src[i]
-		adjustSampleValues(tsSrc.Samples, wr.significantFigures, wr.roundDigits)
-		tssDst = append(tssDst, prompb.TimeSeries{})
-		wr.copyTimeSeries(&tssDst[len(tssDst)-1], tsSrc)
 	}
-
 	wr.tss = tssDst
-	return true
 }
 
-func (wr *writeRequest) copyTimeSeries(dst, src *prompb.TimeSeries) {
-	labelsSrc := src.Labels
-
-	// Pre-allocate memory for labels.
+func (wr *writeRequest) copyTimeSeries(dst, src *prompbmarshal.TimeSeries) {
+	labelsDst := wr.labels
 	labelsLen := len(wr.labels)
-	wr.labels = slicesutil.SetLength(wr.labels, labelsLen+len(labelsSrc))
-	labelsDst := wr.labels[labelsLen:]
+	samplesDst := wr.samples
+	buf := wr.buf
+	for i := range src.Labels {
+		labelsDst = append(labelsDst, prompbmarshal.Label{})
+		dstLabel := &labelsDst[len(labelsDst)-1]
+		srcLabel := &src.Labels[i]
 
-	// Pre-allocate memory for byte slice needed for storing label names and values.
-	neededBufLen := 0
-	for i := range labelsSrc {
-		label := &labelsSrc[i]
-		neededBufLen += len(label.Name) + len(label.Value)
-	}
-	bufLen := len(wr.buf)
-	wr.buf = slicesutil.SetLength(wr.buf, bufLen+neededBufLen)
-	buf := wr.buf[:bufLen]
-
-	// Copy labels
-	for i := range labelsSrc {
-		dstLabel := &labelsDst[i]
-		srcLabel := &labelsSrc[i]
-
-		bufLen := len(buf)
 		buf = append(buf, srcLabel.Name...)
-		dstLabel.Name = bytesutil.ToUnsafeString(buf[bufLen:])
-
-		bufLen = len(buf)
+		dstLabel.Name = bytesutil.ToUnsafeString(buf[len(buf)-len(srcLabel.Name):])
 		buf = append(buf, srcLabel.Value...)
-		dstLabel.Value = bytesutil.ToUnsafeString(buf[bufLen:])
+		dstLabel.Value = bytesutil.ToUnsafeString(buf[len(buf)-len(srcLabel.Value):])
 	}
-	wr.buf = buf
-	dst.Labels = labelsDst
+	dst.Labels = labelsDst[labelsLen:]
 
-	// Copy samples
-	samplesLen := len(wr.samples)
-	wr.samples = append(wr.samples, src.Samples...)
-	dst.Samples = wr.samples[samplesLen:]
+	samplesDst = append(samplesDst, src.Samples...)
+	dst.Samples = samplesDst[len(samplesDst)-len(src.Samples):]
+
+	wr.samples = samplesDst
+	wr.labels = labelsDst
+	wr.buf = buf
 }
 
-// marshalConcurrency limits the maximum number of concurrent workers, which marshal and compress WriteRequest.
-var marshalConcurrencyCh = make(chan struct{}, cgroup.AvailableCPUs())
-
-func tryPushWriteRequest(wr *prompb.WriteRequest, tryPushBlock func(block []byte) bool, isVMRemoteWrite bool) bool {
-	if wr.IsEmpty() {
+func pushWriteRequest(wr *prompbmarshal.WriteRequest, pushBlock func(block []byte), isVMRemoteWrite bool) {
+	if len(wr.Timeseries) == 0 {
 		// Nothing to push
-		return true
+		return
 	}
-
-	marshalConcurrencyCh <- struct{}{}
-
 	bb := writeRequestBufPool.Get()
-	bb.B = wr.MarshalProtobuf(bb.B[:0])
+	bb.B = prompbmarshal.MarshalWriteRequest(bb.B[:0], wr)
 	if len(bb.B) <= maxUnpackedBlockSize.IntN() {
-		zb := compressBufPool.Get()
+		zb := snappyBufPool.Get()
 		if isVMRemoteWrite {
 			zb.B = zstd.CompressLevel(zb.B[:0], bb.B, *vmProtoCompressLevel)
 		} else {
 			zb.B = snappy.Encode(zb.B[:cap(zb.B)], bb.B)
 		}
 		writeRequestBufPool.Put(bb)
-
-		<-marshalConcurrencyCh
-
 		if len(zb.B) <= persistentqueue.MaxBlockSize {
-			zbLen := len(zb.B)
-			ok := tryPushBlock(zb.B)
-			compressBufPool.Put(zb)
-			if ok {
-				blockSizeRows.Update(float64(len(wr.Timeseries)))
-				blockMetadataRows.Update(float64(len(wr.Metadata)))
-				blockSizeBytes.Update(float64(zbLen))
-			}
-			return ok
+			pushBlock(zb.B)
+			blockSizeRows.Update(float64(len(wr.Timeseries)))
+			blockSizeBytes.Update(float64(len(zb.B)))
+			snappyBufPool.Put(zb)
+			return
 		}
-		compressBufPool.Put(zb)
+		snappyBufPool.Put(zb)
 	} else {
 		writeRequestBufPool.Put(bb)
-
-		<-marshalConcurrencyCh
 	}
 
-	// Split timeseries or metadata into two smaller blocks
-	switch len(wr.Timeseries) {
-	case 0:
-		if len(wr.Metadata) == 1 {
-			logger.Warnf("dropping a metadata exceeding -remoteWrite.maxBlockSize=%d bytes", maxUnpackedBlockSize.N)
-			return true
-		}
-		metadata := wr.Metadata
-		n := len(metadata) / 2
-		wr.Metadata = metadata[:n]
-		if !tryPushWriteRequest(wr, tryPushBlock, isVMRemoteWrite) {
-			wr.Metadata = metadata
-			return false
-		}
-		wr.Metadata = metadata[n:]
-		if !tryPushWriteRequest(wr, tryPushBlock, isVMRemoteWrite) {
-			wr.Metadata = metadata
-			return false
-		}
-		wr.Metadata = metadata
-		return true
-
-	case 1:
-		// A single time series left. Recursively split its samples and metadata into smaller parts if possible.
+	// Too big block. Recursively split it into smaller parts if possible.
+	if len(wr.Timeseries) == 1 {
+		// A single time series left. Recursively split its samples into smaller parts if possible.
 		samples := wr.Timeseries[0].Samples
-		metaData := wr.Metadata
-		if len(samples) == 1 && len(metaData) <= 1 {
-			logger.Warnf("dropping a sample for metric and %d metadata which are exceeding -remoteWrite.maxBlockSize=%d bytes", len(metaData), maxUnpackedBlockSize.N)
-			return true
+		if len(samples) == 1 {
+			logger.Warnf("dropping a sample for metric with too long labels exceeding -remoteWrite.maxBlockSize=%d bytes", maxUnpackedBlockSize.N)
+			return
 		}
 		n := len(samples) / 2
-		m := len(metaData) / 2
 		wr.Timeseries[0].Samples = samples[:n]
-		wr.Metadata = metaData[:m]
-		if !tryPushWriteRequest(wr, tryPushBlock, isVMRemoteWrite) {
-			wr.Timeseries[0].Samples = samples
-			wr.Metadata = metaData
-			return false
-		}
+		pushWriteRequest(wr, pushBlock, isVMRemoteWrite)
 		wr.Timeseries[0].Samples = samples[n:]
-		wr.Metadata = metaData[m:]
-		if !tryPushWriteRequest(wr, tryPushBlock, isVMRemoteWrite) {
-			wr.Timeseries[0].Samples = samples
-			wr.Metadata = metaData
-			return false
-		}
+		pushWriteRequest(wr, pushBlock, isVMRemoteWrite)
 		wr.Timeseries[0].Samples = samples
-		wr.Metadata = metaData
-		return true
-
-	default:
-		// Split both timeseries and metadata.
-		timeseries := wr.Timeseries
-		metaData := wr.Metadata
-		n := len(timeseries) / 2
-		m := len(metaData) / 2
-		wr.Timeseries = timeseries[:n]
-		wr.Metadata = metaData[:m]
-		if !tryPushWriteRequest(wr, tryPushBlock, isVMRemoteWrite) {
-			wr.Timeseries = timeseries
-			wr.Metadata = metaData
-			return false
-		}
-		wr.Timeseries = timeseries[n:]
-		wr.Metadata = metaData[m:]
-		if !tryPushWriteRequest(wr, tryPushBlock, isVMRemoteWrite) {
-			wr.Timeseries = timeseries
-			wr.Metadata = metaData
-			return false
-		}
-		wr.Timeseries = timeseries
-		wr.Metadata = metaData
-		return true
+		return
 	}
+	timeseries := wr.Timeseries
+	n := len(timeseries) / 2
+	wr.Timeseries = timeseries[:n]
+	pushWriteRequest(wr, pushBlock, isVMRemoteWrite)
+	wr.Timeseries = timeseries[n:]
+	pushWriteRequest(wr, pushBlock, isVMRemoteWrite)
+	wr.Timeseries = timeseries
 }
 
 var (
-	blockSizeBytes    = metrics.NewHistogram(`vmagent_remotewrite_block_size_bytes`)
-	blockSizeRows     = metrics.NewHistogram(`vmagent_remotewrite_block_size_rows`)
-	blockMetadataRows = metrics.NewHistogram(`vmagent_remotewrite_block_metadata_rows`)
+	blockSizeBytes = metrics.NewHistogram(`vmagent_remotewrite_block_size_bytes`)
+	blockSizeRows  = metrics.NewHistogram(`vmagent_remotewrite_block_size_rows`)
 )
 
-var (
-	writeRequestBufPool bytesutil.ByteBufferPool
-	compressBufPool     bytesutil.ByteBufferPool
-)
+var writeRequestBufPool bytesutil.ByteBufferPool
+var snappyBufPool bytesutil.ByteBufferPool

app/vmagent/remotewrite/pendingseries_test.go

@@ -5,7 +5,7 @@ import (
 	"math"
 	"testing"
 
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
 )
 
 func TestPushWriteRequest(t *testing.T) {
@@ -26,16 +26,13 @@ func testPushWriteRequest(t *testing.T, rowsCount, expectedBlockLenProm, expecte
 		t.Helper()
 		wr := newTestWriteRequest(rowsCount, 20)
 		pushBlockLen := 0
-		pushBlock := func(block []byte) bool {
+		pushBlock := func(block []byte) {
 			if pushBlockLen > 0 {
 				panic(fmt.Errorf("BUG: pushBlock called multiple times; pushBlockLen=%d at first call, len(block)=%d at second call", pushBlockLen, len(block)))
 			}
 			pushBlockLen = len(block)
-			return true
-		}
-		if !tryPushWriteRequest(wr, pushBlock, isVMRemoteWrite) {
-			t.Fatalf("cannot push data to remote storage")
 		}
+		pushWriteRequest(wr, pushBlock, isVMRemoteWrite)
 		if math.Abs(float64(pushBlockLen-expectedBlockLen)/float64(expectedBlockLen)*100) > tolerancePrc {
 			t.Fatalf("unexpected block len for rowsCount=%d, isVMRemoteWrite=%v; got %d bytes; expecting %d bytes +- %.0f%%",
 				rowsCount, isVMRemoteWrite, pushBlockLen, expectedBlockLen, tolerancePrc)
@@ -43,25 +40,25 @@ func testPushWriteRequest(t *testing.T, rowsCount, expectedBlockLenProm, expecte
 	}
 
 	// Check Prometheus remote write
-	f(false, expectedBlockLenProm, 3)
+	f(false, expectedBlockLenProm, 0)
 
 	// Check VictoriaMetrics remote write
 	f(true, expectedBlockLenVM, 15)
 }
 
-func newTestWriteRequest(seriesCount, labelsCount int) *prompb.WriteRequest {
-	var wr prompb.WriteRequest
+func newTestWriteRequest(seriesCount, labelsCount int) *prompbmarshal.WriteRequest {
+	var wr prompbmarshal.WriteRequest
 	for i := 0; i < seriesCount; i++ {
-		var labels []prompb.Label
+		var labels []prompbmarshal.Label
 		for j := 0; j < labelsCount; j++ {
-			labels = append(labels, prompb.Label{
+			labels = append(labels, prompbmarshal.Label{
 				Name:  fmt.Sprintf("label_%d_%d", i, j),
 				Value: fmt.Sprintf("value_%d_%d", i, j),
 			})
 		}
-		wr.Timeseries = append(wr.Timeseries, prompb.TimeSeries{
+		wr.Timeseries = append(wr.Timeseries, prompbmarshal.TimeSeries{
 			Labels: labels,
-			Samples: []prompb.Sample{
+			Samples: []prompbmarshal.Sample{
 				{
 					Value:     float64(i),
 					Timestamp: 1000 * int64(i),

app/vmagent/remotewrite/pendingseries_timing_test.go

@@ -4,6 +4,7 @@ import (
 	"fmt"
 	"testing"
 
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
 	"github.com/golang/snappy"
 	"github.com/klauspost/compress/s2"
 )
@@ -21,7 +22,7 @@ func benchmarkCompressWriteRequest(b *testing.B, compressFunc func(dst, src []by
 	for _, rowsCount := range []int{1, 10, 100, 1e3, 1e4} {
 		b.Run(fmt.Sprintf("rows_%d", rowsCount), func(b *testing.B) {
 			wr := newTestWriteRequest(rowsCount, 10)
-			data := wr.MarshalProtobuf(nil)
+			data := prompbmarshal.MarshalWriteRequest(nil, wr)
 			b.ReportAllocs()
 			b.SetBytes(int64(rowsCount))
 			b.RunParallel(func(pb *testing.PB) {

app/vmagent/remotewrite/relabel.go

@@ -3,17 +3,13 @@ package remotewrite
 import (
 	"flag"
 	"fmt"
-	"strconv"
 	"strings"
 	"sync"
 
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/fasttime"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promrelabel"
-
-	"github.com/VictoriaMetrics/metrics"
 )
 
 var (
@@ -22,31 +18,17 @@ var (
 	relabelConfigPathGlobal = flag.String("remoteWrite.relabelConfig", "", "Optional path to file with relabeling configs, which are applied "+
 		"to all the metrics before sending them to -remoteWrite.url. See also -remoteWrite.urlRelabelConfig. "+
 		"The path can point either to local file or to http url. "+
-		"See https://docs.victoriametrics.com/victoriametrics/relabeling/")
+		"See https://docs.victoriametrics.com/vmagent.html#relabeling")
 	relabelConfigPaths = flagutil.NewArrayString("remoteWrite.urlRelabelConfig", "Optional path to relabel configs for the corresponding -remoteWrite.url. "+
 		"See also -remoteWrite.relabelConfig. The path can point either to local file or to http url. "+
-		"See https://docs.victoriametrics.com/victoriametrics/relabeling/")
+		"See https://docs.victoriametrics.com/vmagent.html#relabeling")
 
 	usePromCompatibleNaming = flag.Bool("usePromCompatibleNaming", false, "Whether to replace characters unsupported by Prometheus with underscores "+
 		"in the ingested metric names and label names. For example, foo.bar{a.b='c'} is transformed into foo_bar{a_b='c'} during data ingestion if this flag is set. "+
 		"See https://prometheus.io/docs/concepts/data_model/#metric-names-and-labels")
 )
 
-var labelsGlobal []prompb.Label
-
-var (
-	relabelConfigReloads      *metrics.Counter
-	relabelConfigReloadErrors *metrics.Counter
-	relabelConfigSuccess      *metrics.Gauge
-	relabelConfigTimestamp    *metrics.Counter
-)
-
-func initRelabelMetrics() {
-	relabelConfigReloads = metrics.NewCounter(`vmagent_relabel_config_reloads_total`)
-	relabelConfigReloadErrors = metrics.NewCounter(`vmagent_relabel_config_reloads_errors_total`)
-	relabelConfigSuccess = metrics.NewGauge(`vmagent_relabel_config_last_reload_successful`, nil)
-	relabelConfigTimestamp = metrics.NewCounter(`vmagent_relabel_config_last_reload_success_timestamp_seconds`)
-}
+var labelsGlobal []prompbmarshal.Label
 
 // CheckRelabelConfigs checks -remoteWrite.relabelConfig and -remoteWrite.urlRelabelConfig.
 func CheckRelabelConfigs() error {
@@ -54,39 +36,6 @@ func CheckRelabelConfigs() error {
 	return err
 }
 
-func initRelabelConfigs() {
-	rcs, err := loadRelabelConfigs()
-	if err != nil {
-		logger.Fatalf("cannot initialize relabel configs: %s", err)
-	}
-	allRelabelConfigs.Store(rcs)
-	if rcs.isSet() {
-		initRelabelMetrics()
-		relabelConfigSuccess.Set(1)
-		relabelConfigTimestamp.Set(fasttime.UnixTimestamp())
-	}
-}
-
-func reloadRelabelConfigs() {
-	rcs := allRelabelConfigs.Load()
-	if !rcs.isSet() {
-		return
-	}
-	relabelConfigReloads.Inc()
-	logger.Infof("reloading relabel configs pointed by -remoteWrite.relabelConfig and -remoteWrite.urlRelabelConfig")
-	rcs, err := loadRelabelConfigs()
-	if err != nil {
-		relabelConfigReloadErrors.Inc()
-		relabelConfigSuccess.Set(0)
-		logger.Errorf("cannot reload relabel configs; preserving the previous configs; error: %s", err)
-		return
-	}
-	allRelabelConfigs.Store(rcs)
-	relabelConfigSuccess.Set(1)
-	relabelConfigTimestamp.Set(fasttime.UnixTimestamp())
-	logger.Infof("successfully reloaded relabel configs")
-}
-
 func loadRelabelConfigs() (*relabelConfigs, error) {
 	var rcs relabelConfigs
 	if *relabelConfigPathGlobal != "" {
@@ -96,11 +45,11 @@ func loadRelabelConfigs() (*relabelConfigs, error) {
 		}
 		rcs.global = global
 	}
-	if len(*relabelConfigPaths) > len(*remoteWriteURLs) {
-		return nil, fmt.Errorf("too many -remoteWrite.urlRelabelConfig args: %d; it mustn't exceed the number of -remoteWrite.url args: %d",
-			len(*relabelConfigPaths), (len(*remoteWriteURLs)))
+	if len(*relabelConfigPaths) > (len(*remoteWriteURLs) + len(*remoteWriteMultitenantURLs)) {
+		return nil, fmt.Errorf("too many -remoteWrite.urlRelabelConfig args: %d; it mustn't exceed the number of -remoteWrite.url or -remoteWrite.multitenantURL args: %d",
+			len(*relabelConfigPaths), (len(*remoteWriteURLs) + len(*remoteWriteMultitenantURLs)))
 	}
-	rcs.perURL = make([]*promrelabel.ParsedConfigs, len(*remoteWriteURLs))
+	rcs.perURL = make([]*promrelabel.ParsedConfigs, (len(*remoteWriteURLs) + len(*remoteWriteMultitenantURLs)))
 	for i, path := range *relabelConfigPaths {
 		if len(path) == 0 {
 			// Skip empty relabel config.
@@ -120,21 +69,6 @@ type relabelConfigs struct {
 	perURL []*promrelabel.ParsedConfigs
 }
 
-func (rcs *relabelConfigs) isSet() bool {
-	if rcs == nil {
-		return false
-	}
-	if rcs.global.Len() > 0 {
-		return true
-	}
-	for _, pc := range rcs.perURL {
-		if pc.Len() > 0 {
-			return true
-		}
-	}
-	return false
-}
-
 // initLabelsGlobal must be called after parsing command-line flags.
 func initLabelsGlobal() {
 	labelsGlobal = nil
@@ -146,35 +80,53 @@ func initLabelsGlobal() {
 		if n < 0 {
 			logger.Fatalf("missing '=' in `-remoteWrite.label`. It must contain label in the form `name=value`; got %q", s)
 		}
-		labelsGlobal = append(labelsGlobal, prompb.Label{
+		labelsGlobal = append(labelsGlobal, prompbmarshal.Label{
 			Name:  s[:n],
 			Value: s[n+1:],
 		})
 	}
 }
 
-func (rctx *relabelCtx) applyRelabeling(tss []prompb.TimeSeries, pcs *promrelabel.ParsedConfigs) []prompb.TimeSeries {
-	if pcs.Len() == 0 && !*usePromCompatibleNaming {
+func (rctx *relabelCtx) applyRelabeling(tss []prompbmarshal.TimeSeries, extraLabels []prompbmarshal.Label, pcs *promrelabel.ParsedConfigs) []prompbmarshal.TimeSeries {
+	if len(extraLabels) == 0 && pcs.Len() == 0 && !*usePromCompatibleNaming {
 		// Nothing to change.
 		return tss
 	}
-	rctx.reset()
 	tssDst := tss[:0]
 	labels := rctx.labels[:0]
 	for i := range tss {
 		ts := &tss[i]
 		labelsLen := len(labels)
 		labels = append(labels, ts.Labels...)
+		// extraLabels must be added before applying relabeling according to https://prometheus.io/docs/prometheus/latest/configuration/configuration/#remote_write
+		for j := range extraLabels {
+			extraLabel := &extraLabels[j]
+			tmp := promrelabel.GetLabelByName(labels[labelsLen:], extraLabel.Name)
+			if tmp != nil {
+				tmp.Value = extraLabel.Value
+			} else {
+				labels = append(labels, *extraLabel)
+			}
+		}
+		if *usePromCompatibleNaming {
+			// Replace unsupported Prometheus chars in label names and metric names with underscores.
+			tmpLabels := labels[labelsLen:]
+			for j := range tmpLabels {
+				label := &tmpLabels[j]
+				if label.Name == "__name__" {
+					label.Value = promrelabel.SanitizeName(label.Value)
+				} else {
+					label.Name = promrelabel.SanitizeName(label.Name)
+				}
+			}
+		}
 		labels = pcs.Apply(labels, labelsLen)
 		labels = promrelabel.FinalizeLabels(labels[:labelsLen], labels[labelsLen:])
 		if len(labels) == labelsLen {
 			// Drop the current time series, since relabeling removed all the labels.
 			continue
 		}
-		if *usePromCompatibleNaming {
-			fixPromCompatibleNaming(labels[labelsLen:])
-		}
-		tssDst = append(tssDst, prompb.TimeSeries{
+		tssDst = append(tssDst, prompbmarshal.TimeSeries{
 			Labels:  labels[labelsLen:],
 			Samples: ts.Samples,
 		})
@@ -183,61 +135,9 @@ func (rctx *relabelCtx) applyRelabeling(tss []prompb.TimeSeries, pcs *promrelabe
 	return tssDst
 }
 
-func (rctx *relabelCtx) appendExtraLabels(tss []prompb.TimeSeries, extraLabels []prompb.Label) {
-	if len(extraLabels) == 0 {
-		return
-	}
-	rctx.reset()
-	labels := rctx.labels[:0]
-	for i := range tss {
-		ts := &tss[i]
-		labelsLen := len(labels)
-		labels = append(labels, ts.Labels...)
-		for j := range extraLabels {
-			extraLabel := extraLabels[j]
-			tmp := promrelabel.GetLabelByName(labels[labelsLen:], extraLabel.Name)
-			if tmp != nil {
-				tmp.Value = extraLabel.Value
-			} else {
-				labels = append(labels, extraLabel)
-			}
-		}
-		ts.Labels = labels[labelsLen:]
-	}
-	rctx.labels = labels
-}
-
-func (rctx *relabelCtx) tenantToLabels(tss []prompb.TimeSeries, accountID, projectID uint32) {
-	rctx.reset()
-	accountIDStr := strconv.FormatUint(uint64(accountID), 10)
-	projectIDStr := strconv.FormatUint(uint64(projectID), 10)
-	labels := rctx.labels[:0]
-	for i := range tss {
-		ts := &tss[i]
-		labelsLen := len(labels)
-		for _, label := range ts.Labels {
-			labelName := label.Name
-			if labelName == "vm_account_id" || labelName == "vm_project_id" {
-				continue
-			}
-			labels = append(labels, label)
-		}
-		labels = append(labels, prompb.Label{
-			Name:  "vm_account_id",
-			Value: accountIDStr,
-		})
-		labels = append(labels, prompb.Label{
-			Name:  "vm_project_id",
-			Value: projectIDStr,
-		})
-		ts.Labels = labels[labelsLen:]
-	}
-	rctx.labels = labels
-}
-
 type relabelCtx struct {
 	// pool for labels, which are used during the relabeling.
-	labels []prompb.Label
+	labels []prompbmarshal.Label
 }
 
 func (rctx *relabelCtx) reset() {
@@ -246,7 +146,7 @@ func (rctx *relabelCtx) reset() {
 }
 
 var relabelCtxPool = &sync.Pool{
-	New: func() any {
+	New: func() interface{} {
 		return &relabelCtx{}
 	},
 }
@@ -256,18 +156,6 @@ func getRelabelCtx() *relabelCtx {
 }
 
 func putRelabelCtx(rctx *relabelCtx) {
-	rctx.reset()
+	rctx.labels = rctx.labels[:0]
 	relabelCtxPool.Put(rctx)
 }
-
-func fixPromCompatibleNaming(labels []prompb.Label) {
-	// Replace unsupported Prometheus chars in label names and metric names with underscores.
-	for i := range labels {
-		label := &labels[i]
-		if label.Name == "__name__" {
-			label.Value = promrelabel.SanitizeMetricName(label.Value)
-		} else {
-			label.Name = promrelabel.SanitizeLabelName(label.Name)
-		}
-	}
-}

app/vmagent/remotewrite/relabel_test.go

@@ -4,22 +4,24 @@ import (
 	"reflect"
 	"testing"
 
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promrelabel"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promutil"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promutils"
 )
 
 func TestApplyRelabeling(t *testing.T) {
-	f := func(pcs *promrelabel.ParsedConfigs, sTss, sExpTss string) {
+	f := func(extraLabels []prompbmarshal.Label, pcs *promrelabel.ParsedConfigs, sTss, sExpTss string) {
 		rctx := &relabelCtx{}
 		tss, expTss := parseSeries(sTss), parseSeries(sExpTss)
-		gotTss := rctx.applyRelabeling(tss, pcs)
+		gotTss := rctx.applyRelabeling(tss, extraLabels, pcs)
 		if !reflect.DeepEqual(gotTss, expTss) {
 			t.Fatalf("expected to have: \n%v;\ngot: \n%v", expTss, gotTss)
 		}
 	}
 
-	f(nil, "up", "up")
+	f(nil, nil, "up", "up")
+	f([]prompbmarshal.Label{{Name: "foo", Value: "bar"}}, nil, "up", `up{foo="bar"}`)
+	f([]prompbmarshal.Label{{Name: "foo", Value: "bar"}}, nil, `up{foo="baz"}`, `up{foo="bar"}`)
 
 	pcs, err := promrelabel.ParseRelabelConfigsData([]byte(`
 - target_label: "foo"
@@ -30,40 +32,18 @@ func TestApplyRelabeling(t *testing.T) {
 	if err != nil {
 		t.Fatalf("unexpected error: %s", err)
 	}
-	f(pcs, `up{foo="baz", env="prod"}`, `up{foo="aaa"}`)
+	f(nil, pcs, `up{foo="baz", env="prod"}`, `up{foo="aaa"}`)
 
 	oldVal := *usePromCompatibleNaming
 	*usePromCompatibleNaming = true
-	f(nil, `foo.bar`, `foo_bar`)
+	f(nil, nil, `foo.bar`, `foo_bar`)
 	*usePromCompatibleNaming = oldVal
 }
 
-func TestAppendExtraLabels(t *testing.T) {
-	f := func(extraLabels []prompb.Label, sTss, sExpTss string) {
-		t.Helper()
-		rctx := &relabelCtx{}
-		tss, expTss := parseSeries(sTss), parseSeries(sExpTss)
-		rctx.appendExtraLabels(tss, extraLabels)
-		if !reflect.DeepEqual(tss, expTss) {
-			t.Fatalf("expected to have: \n%v;\ngot: \n%v", expTss, tss)
-		}
-	}
-
-	f(nil, "up", "up")
-	f([]prompb.Label{{Name: "foo", Value: "bar"}}, "up", `up{foo="bar"}`)
-	f([]prompb.Label{{Name: "foo", Value: "bar"}}, `up{foo="baz"}`, `up{foo="bar"}`)
-	f([]prompb.Label{{Name: "baz", Value: "qux"}}, `up{foo="baz"}`, `up{foo="baz",baz="qux"}`)
-
-	oldVal := *usePromCompatibleNaming
-	*usePromCompatibleNaming = true
-	f([]prompb.Label{{Name: "foo.bar", Value: "baz"}}, "up", `up{foo.bar="baz"}`)
-	*usePromCompatibleNaming = oldVal
-}
-
-func parseSeries(data string) []prompb.TimeSeries {
-	var tss []prompb.TimeSeries
-	tss = append(tss, prompb.TimeSeries{
-		Labels: promutil.MustNewLabelsFromString(data).GetLabels(),
+func parseSeries(data string) []prompbmarshal.TimeSeries {
+	var tss []prompbmarshal.TimeSeries
+	tss = append(tss, prompbmarshal.TimeSeries{
+		Labels: promutils.MustNewLabelsFromString(data).GetLabels(),
 	})
 	return tss
 }

app/vmagent/remotewrite/remotewrite_test.go

@@ -1,349 +0,0 @@
-package remotewrite
-
-import (
-	"fmt"
-	"math"
-	"reflect"
-	"strconv"
-	"sync/atomic"
-	"testing"
-	"time"
-
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/consistenthash"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promrelabel"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/prometheus"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/streamaggr"
-
-	"github.com/VictoriaMetrics/metrics"
-)
-
-func TestGetLabelsHash_Distribution(t *testing.T) {
-	f := func(bucketsCount int) {
-		t.Helper()
-
-		// Distribute itemsCount hashes returned by getLabelsHash() across bucketsCount buckets.
-		itemsCount := 1_000 * bucketsCount
-		m := make([]int, bucketsCount)
-		var labels []prompb.Label
-		for i := 0; i < itemsCount; i++ {
-			labels = append(labels[:0], prompb.Label{
-				Name:  "__name__",
-				Value: fmt.Sprintf("some_name_%d", i),
-			})
-			for j := 0; j < 10; j++ {
-				labels = append(labels, prompb.Label{
-					Name:  fmt.Sprintf("label_%d", j),
-					Value: fmt.Sprintf("value_%d_%d", i, j),
-				})
-			}
-			h := getLabelsHash(labels)
-			m[h%uint64(bucketsCount)]++
-		}
-
-		// Verify that the distribution is even
-		expectedItemsPerBucket := itemsCount / bucketsCount
-		for _, n := range m {
-			if math.Abs(1-float64(n)/float64(expectedItemsPerBucket)) > 0.04 {
-				t.Fatalf("unexpected items in the bucket for %d buckets; got %d; want around %d", bucketsCount, n, expectedItemsPerBucket)
-			}
-		}
-	}
-
-	f(2)
-	f(3)
-	f(4)
-	f(5)
-	f(10)
-}
-
-func TestRemoteWriteContext_TryPush_ImmutableTimeseries(t *testing.T) {
-	f := func(streamAggrConfig, relabelConfig string, enableWindows bool, dedupInterval time.Duration, keepInput, dropInput bool, input string) {
-		t.Helper()
-		perURLRelabel, err := promrelabel.ParseRelabelConfigsData([]byte(relabelConfig))
-		if err != nil {
-			t.Fatalf("cannot load relabel configs: %s", err)
-		}
-		rcs := &relabelConfigs{
-			perURL: []*promrelabel.ParsedConfigs{
-				perURLRelabel,
-			},
-		}
-		allRelabelConfigs.Store(rcs)
-
-		pss := make([]*pendingSeries, 1)
-		isVMProto := &atomic.Bool{}
-		isVMProto.Store(true)
-		pss[0] = newPendingSeries(nil, isVMProto, 0, 100)
-		rwctx := &remoteWriteCtx{
-			idx:                    0,
-			streamAggrKeepInput:    keepInput,
-			streamAggrDropInput:    dropInput,
-			pss:                    pss,
-			rowsPushedAfterRelabel: metrics.GetOrCreateCounter(`foo`),
-			rowsDroppedByRelabel:   metrics.GetOrCreateCounter(`bar`),
-		}
-		if dedupInterval > 0 {
-			rwctx.deduplicator = streamaggr.NewDeduplicator(nil, enableWindows, dedupInterval, nil, "dedup-global")
-		}
-
-		if streamAggrConfig != "" {
-			pushNoop := func(_ []prompb.TimeSeries) {}
-			opts := streamaggr.Options{
-				EnableWindows: enableWindows,
-			}
-			sas, err := streamaggr.LoadFromData([]byte(streamAggrConfig), pushNoop, &opts, "global")
-			if err != nil {
-				t.Fatalf("cannot load streamaggr configs: %s", err)
-			}
-			defer sas.MustStop()
-			rwctx.sas.Store(sas)
-		}
-
-		offsetMsecs := time.Now().UnixMilli()
-		inputTss := prometheus.MustParsePromMetrics(input, offsetMsecs)
-		expectedTss := make([]prompb.TimeSeries, len(inputTss))
-
-		// copy inputTss to make sure it is not mutated during TryPush call
-		copy(expectedTss, inputTss)
-		if !rwctx.TryPushTimeSeries(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)
-		}
-	}
-
-	f(`
-- interval: 1m
-  outputs: [sum_samples]
-- interval: 2m
-  outputs: [count_series]
-`, `
-- action: keep
-  source_labels: [env]
-  regex: "dev"
-`, false, 0, false, false, `
-metric{env="dev"} 10
-metric{env="bar"} 20
-metric{env="dev"} 15
-metric{env="bar"} 25
-`)
-	f(``, ``, true, time.Hour, false, false, `
-metric{env="dev"} 10
-metric{env="foo"} 20
-metric{env="dev"} 15
-metric{env="foo"} 25
-`)
-	f(``, `
-- action: keep
-  source_labels: [env]
-  regex: "dev"
-`, true, time.Hour, false, false, `
-metric{env="dev"} 10
-metric{env="bar"} 20
-metric{env="dev"} 15
-metric{env="bar"} 25
-`)
-	f(``, `
-- action: keep
-  source_labels: [env]
-  regex: "dev"
-`, true, time.Hour, true, false, `
-metric{env="test"} 10
-metric{env="dev"} 20
-metric{env="foo"} 15
-metric{env="dev"} 25
-`)
-	f(``, `
-- action: keep
-  source_labels: [env]
-  regex: "dev"
-`, true, time.Hour, false, true, `
-metric{env="foo"} 10
-metric{env="dev"} 20
-metric{env="foo"} 15
-metric{env="dev"} 25
-`)
-	f(``, `
-- action: keep
-  source_labels: [env]
-  regex: "dev"
-`, true, time.Hour, true, true, `
-metric{env="dev"} 10
-metric{env="test"} 20
-metric{env="dev"} 15
-metric{env="bar"} 25
-`)
-}
-
-func TestShardAmountRemoteWriteCtx(t *testing.T) {
-	// 1. distribute 100000 series to n nodes.
-	// 2. remove the last node from healthy list.
-	// 3. distribute the same 10000 series to (n-1) node again.
-	// 4. check active time series change rate:
-	// change rate must < (3/total nodes). e.g. +30% if 10 you have 10 nodes.
-
-	f := func(remoteWriteCount int, healthyIdx []int, replicas int) {
-		t.Helper()
-		defer func() {
-			rwctxsGlobal = nil
-			rwctxsGlobalIdx = nil
-			rwctxConsistentHashGlobal = nil
-		}()
-
-		rwctxsGlobal = make([]*remoteWriteCtx, remoteWriteCount)
-		rwctxsGlobalIdx = make([]int, remoteWriteCount)
-		rwctxs := make([]*remoteWriteCtx, 0, len(healthyIdx))
-
-		for i := range remoteWriteCount {
-			rwCtx := &remoteWriteCtx{
-				idx: i,
-			}
-			rwctxsGlobalIdx[i] = i
-
-			if i >= len(healthyIdx) {
-				rwctxsGlobal[i] = rwCtx
-				continue
-			}
-			hIdx := healthyIdx[i]
-			if hIdx != i {
-				rwctxs = append(rwctxs, &remoteWriteCtx{
-					idx: hIdx,
-				})
-			} else {
-				rwctxs = append(rwctxs, rwCtx)
-			}
-			rwctxsGlobal[i] = rwCtx
-		}
-
-		seriesCount := 100000
-		// build 1000000 series
-		tssBlock := make([]prompb.TimeSeries, 0, seriesCount)
-		for i := 0; i < seriesCount; i++ {
-			tssBlock = append(tssBlock, prompb.TimeSeries{
-				Labels: []prompb.Label{
-					{
-						Name:  "label",
-						Value: strconv.Itoa(i),
-					},
-				},
-				Samples: []prompb.Sample{
-					{
-						Timestamp: 0,
-						Value:     0,
-					},
-				},
-			})
-		}
-
-		// build consistent hash for x remote write context
-		// build active time series set
-		nodes := make([]string, 0, remoteWriteCount)
-		activeTimeSeriesByNodes := make([]map[string]struct{}, remoteWriteCount)
-		for i := 0; i < remoteWriteCount; i++ {
-			nodes = append(nodes, fmt.Sprintf("node%d", i))
-			activeTimeSeriesByNodes[i] = make(map[string]struct{})
-		}
-		rwctxConsistentHashGlobal = consistenthash.NewConsistentHash(nodes, 0)
-
-		// create shards
-		x := getTSSShards(len(rwctxs))
-		shards := x.shards
-
-		// execute
-		shardAmountRemoteWriteCtx(tssBlock, shards, rwctxs, replicas)
-
-		for i, nodeIdx := range healthyIdx {
-			for _, ts := range shards[i] {
-				// add it to node[nodeIdx]'s active time series
-				activeTimeSeriesByNodes[nodeIdx][prompb.LabelsToString(ts.Labels)] = struct{}{}
-			}
-		}
-
-		totalActiveTimeSeries := 0
-		for _, activeTimeSeries := range activeTimeSeriesByNodes {
-			totalActiveTimeSeries += len(activeTimeSeries)
-		}
-		avgActiveTimeSeries1 := totalActiveTimeSeries / remoteWriteCount
-		putTSSShards(x)
-
-		// removed last node
-		rwctxs = rwctxs[:len(rwctxs)-1]
-		healthyIdx = healthyIdx[:len(healthyIdx)-1]
-
-		x = getTSSShards(len(rwctxs))
-		shards = x.shards
-
-		// execute
-		shardAmountRemoteWriteCtx(tssBlock, shards, rwctxs, replicas)
-		for i, nodeIdx := range healthyIdx {
-			for _, ts := range shards[i] {
-				// add it to node[nodeIdx]'s active time series
-				activeTimeSeriesByNodes[nodeIdx][prompb.LabelsToString(ts.Labels)] = struct{}{}
-			}
-		}
-
-		totalActiveTimeSeries = 0
-		for _, activeTimeSeries := range activeTimeSeriesByNodes {
-			totalActiveTimeSeries += len(activeTimeSeries)
-		}
-		avgActiveTimeSeries2 := totalActiveTimeSeries / remoteWriteCount
-
-		changed := math.Abs(float64(avgActiveTimeSeries2-avgActiveTimeSeries1) / float64(avgActiveTimeSeries1))
-		threshold := 3 / float64(remoteWriteCount)
-
-		if changed >= threshold {
-			t.Fatalf("average active time series before: %d, after: %d, changed: %.2f. threshold: %.2f", avgActiveTimeSeries1, avgActiveTimeSeries2, changed, threshold)
-		}
-
-	}
-
-	f(5, []int{0, 1, 2, 3, 4}, 1)
-
-	f(5, []int{0, 1, 2, 3, 4}, 2)
-
-	f(10, []int{0, 1, 2, 3, 4, 5, 6, 7, 9}, 1)
-
-	f(10, []int{0, 1, 2, 3, 4, 5, 6, 7, 9}, 3)
-}
-
-func TestCalculateHealthyRwctxIdx(t *testing.T) {
-	f := func(total int, healthyIdx []int, unhealthyIdx []int) {
-		t.Helper()
-
-		healthyMap := make(map[int]bool)
-		for _, idx := range healthyIdx {
-			healthyMap[idx] = true
-		}
-		rwctxsGlobal = make([]*remoteWriteCtx, total)
-		rwctxsGlobalIdx = make([]int, total)
-		rwctxs := make([]*remoteWriteCtx, 0, len(healthyIdx))
-		for i := range rwctxsGlobal {
-			rwctx := &remoteWriteCtx{idx: i}
-			rwctxsGlobal[i] = rwctx
-			if healthyMap[i] {
-				rwctxs = append(rwctxs, rwctx)
-			}
-			rwctxsGlobalIdx[i] = i
-		}
-
-		gotHealthyIdx, gotUnhealthyIdx := calculateHealthyRwctxIdx(rwctxs)
-		if !reflect.DeepEqual(healthyIdx, gotHealthyIdx) {
-			t.Errorf("calculateHealthyRwctxIdx want healthyIdx = %v, got %v", healthyIdx, gotHealthyIdx)
-		}
-		if !reflect.DeepEqual(unhealthyIdx, gotUnhealthyIdx) {
-			t.Errorf("calculateHealthyRwctxIdx want unhealthyIdx = %v, got %v", unhealthyIdx, gotUnhealthyIdx)
-		}
-	}
-
-	f(5, []int{0, 1, 2, 3, 4}, nil)
-	f(5, []int{0, 1, 2, 4}, []int{3})
-	f(5, []int{2, 4}, []int{0, 1, 3})
-	f(5, []int{0, 2, 4}, []int{1, 3})
-	f(5, []int{}, []int{0, 1, 2, 3, 4})
-	f(5, []int{4}, []int{0, 1, 2, 3})
-	f(1, []int{0}, nil)
-	f(1, []int{}, []int{0})
-}

app/vmagent/remotewrite/statconn.go

@@ -0,0 +1,92 @@
+package remotewrite
+
+import (
+	"context"
+	"net"
+	"sync"
+	"sync/atomic"
+	"time"
+
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/netutil"
+	"github.com/VictoriaMetrics/metrics"
+)
+
+func getStdDialer() *net.Dialer {
+	stdDialerOnce.Do(func() {
+		stdDialer = &net.Dialer{
+			Timeout:   30 * time.Second,
+			KeepAlive: 30 * time.Second,
+			DualStack: netutil.TCP6Enabled(),
+		}
+	})
+	return stdDialer
+}
+
+var (
+	stdDialer     *net.Dialer
+	stdDialerOnce sync.Once
+)
+
+func statDial(ctx context.Context, networkUnused, addr string) (conn net.Conn, err error) {
+	network := netutil.GetTCPNetwork()
+	d := getStdDialer()
+	conn, err = d.DialContext(ctx, network, addr)
+	dialsTotal.Inc()
+	if err != nil {
+		dialErrors.Inc()
+		return nil, err
+	}
+	conns.Inc()
+	sc := &statConn{
+		Conn: conn,
+	}
+	return sc, nil
+}
+
+var (
+	dialsTotal = metrics.NewCounter(`vmagent_remotewrite_dials_total`)
+	dialErrors = metrics.NewCounter(`vmagent_remotewrite_dial_errors_total`)
+	conns      = metrics.NewCounter(`vmagent_remotewrite_conns`)
+)
+
+type statConn struct {
+	closed uint64
+	net.Conn
+}
+
+func (sc *statConn) Read(p []byte) (int, error) {
+	n, err := sc.Conn.Read(p)
+	connReadsTotal.Inc()
+	if err != nil {
+		connReadErrors.Inc()
+	}
+	connBytesRead.Add(n)
+	return n, err
+}
+
+func (sc *statConn) Write(p []byte) (int, error) {
+	n, err := sc.Conn.Write(p)
+	connWritesTotal.Inc()
+	if err != nil {
+		connWriteErrors.Inc()
+	}
+	connBytesWritten.Add(n)
+	return n, err
+}
+
+func (sc *statConn) Close() error {
+	err := sc.Conn.Close()
+	if atomic.AddUint64(&sc.closed, 1) == 1 {
+		conns.Dec()
+	}
+	return err
+}
+
+var (
+	connReadsTotal   = metrics.NewCounter(`vmagent_remotewrite_conn_reads_total`)
+	connWritesTotal  = metrics.NewCounter(`vmagent_remotewrite_conn_writes_total`)
+	connReadErrors   = metrics.NewCounter(`vmagent_remotewrite_conn_read_errors_total`)
+	connWriteErrors  = metrics.NewCounter(`vmagent_remotewrite_conn_write_errors_total`)
+	connBytesRead    = metrics.NewCounter(`vmagent_remotewrite_conn_bytes_read_total`)
+	connBytesWritten = metrics.NewCounter(`vmagent_remotewrite_conn_bytes_written_total`)
+)

app/vmagent/remotewrite/streamaggr.go

@@ -1,258 +0,0 @@
-package remotewrite
-
-import (
-	"flag"
-	"fmt"
-	"strings"
-
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/fasttime"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/streamaggr"
-	"github.com/VictoriaMetrics/metrics"
-)
-
-var (
-	// Global config
-	streamAggrGlobalConfig = flag.String("streamAggr.config", "", "Optional path to file with stream aggregation config. "+
-		"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/ . "+
-		"See also -streamAggr.keepInput, -streamAggr.dropInput and -streamAggr.dedupInterval")
-	streamAggrGlobalKeepInput = flag.Bool("streamAggr.keepInput", false, "Whether to keep all the input samples after the aggregation "+
-		"with -streamAggr.config. By default, only aggregates samples are dropped, while the remaining samples "+
-		"are written to remote storages write. See also -streamAggr.dropInput and https://docs.victoriametrics.com/victoriametrics/stream-aggregation/")
-	streamAggrGlobalDropInput = flag.Bool("streamAggr.dropInput", false, "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 "+
-		"are written to remote storages write. See also -streamAggr.keepInput and https://docs.victoriametrics.com/victoriametrics/stream-aggregation/")
-	streamAggrGlobalDedupInterval = flag.Duration("streamAggr.dedupInterval", 0, "Input samples are de-duplicated with this interval on "+
-		"aggregator before optional aggregation with -streamAggr.config . "+
-		"See also -dedup.minScrapeInterval and https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#deduplication")
-	streamAggrGlobalIgnoreOldSamples = flag.Bool("streamAggr.ignoreOldSamples", false, "Whether to ignore input samples with old timestamps outside the "+
-		"current aggregation interval for aggregator. "+
-		"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#ignoring-old-samples")
-	streamAggrGlobalIgnoreFirstIntervals = flag.Int("streamAggr.ignoreFirstIntervals", 0, "Number of aggregation intervals to skip after the start for "+
-		"aggregator. 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. See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#ignore-aggregation-intervals-on-start")
-	streamAggrGlobalDropInputLabels = flagutil.NewArrayString("streamAggr.dropInputLabels", "An optional list of labels to drop from samples for aggregator "+
-		"before stream de-duplication and aggregation . See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#dropping-unneeded-labels")
-	streamAggrGlobalEnableWindows = flag.Bool("streamAggr.enableWindows", false, "Enables aggregation within fixed windows for all global aggregators. "+
-		"This allows to get more precise results, but impacts resource usage as it requires twice more memory to store two states. "+
-		"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#aggregation-windows.")
-
-	// Per URL 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/victoriametrics/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 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/victoriametrics/stream-aggregation/")
-	streamAggrKeepInput = flagutil.NewArrayBool("remoteWrite.streamAggr.keepInput", "Whether to keep all the input samples after the aggregation "+
-		"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/victoriametrics/stream-aggregation/")
-	streamAggrDedupInterval = flagutil.NewArrayDuration("remoteWrite.streamAggr.dedupInterval", 0, "Input samples are de-duplicated with this interval before optional aggregation "+
-		"with -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. See also -dedup.minScrapeInterval and https://docs.victoriametrics.com/victoriametrics/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 at the corresponding -remoteWrite.url. "+
-		"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#ignoring-old-samples")
-	streamAggrIgnoreFirstIntervals = flagutil.NewArrayInt("remoteWrite.streamAggr.ignoreFirstIntervals", 0, "Number of aggregation intervals to skip after the start "+
-		"for the corresponding -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. Increase this value if "+
-		"you observe incorrect aggregation results after vmagent restarts. It could be caused by receiving buffered delayed data from clients pushing data into the vmagent. "+
-		"See https://docs.victoriametrics.com/victoriametrics/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 with -remoteWrite.streamAggr.config and -remoteWrite.streamAggr.dedupInterval at the corresponding -remoteWrite.url. "+
-		"Multiple labels per remoteWrite.url must be delimited by '^^': -remoteWrite.streamAggr.dropInputLabels='replica^^az,replica'. "+
-		"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#dropping-unneeded-labels")
-	streamAggrEnableWindows = flagutil.NewArrayBool("remoteWrite.streamAggr.enableWindows", "Enables aggregation within fixed windows for all remote write's aggregators. "+
-		"This allows to get more precise results, but impacts resource usage as it requires twice more memory to store two states. "+
-		"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#aggregation-windows.")
-)
-
-// CheckStreamAggrConfigs checks -remoteWrite.streamAggr.config and -streamAggr.config.
-func CheckStreamAggrConfigs() error {
-	// Check global config
-	sas, err := newStreamAggrConfigGlobal()
-	if err != nil {
-		return err
-	}
-	sas.MustStop()
-
-	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))
-	}
-
-	pushNoop := func(_ []prompb.TimeSeries) {}
-	for idx := range *streamAggrConfig {
-		sas, err := newStreamAggrConfigPerURL(idx, pushNoop)
-		if err != nil {
-			return err
-		}
-		sas.MustStop()
-	}
-	return nil
-}
-
-func reloadStreamAggrConfigs() {
-	reloadStreamAggrConfigGlobal()
-	for _, rwctx := range rwctxsGlobal {
-		rwctx.reloadStreamAggrConfig()
-	}
-}
-
-func reloadStreamAggrConfigGlobal() {
-	path := *streamAggrGlobalConfig
-	if path == "" {
-		return
-	}
-
-	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) {
-		sasOld := sasGlobal.Swap(sasNew)
-		sasOld.MustStop()
-		logger.Infof("successfully reloaded -streamAggr.config=%q", path)
-	} else {
-		sasNew.MustStop()
-		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 initStreamAggrConfigGlobal() {
-	sas, err := newStreamAggrConfigGlobal()
-	if err != nil {
-		logger.Fatalf("cannot initialize global 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())
-	}
-	dedupInterval := *streamAggrGlobalDedupInterval
-	if dedupInterval > 0 {
-		deduplicatorGlobal = streamaggr.NewDeduplicator(pushTimeSeriesToRemoteStoragesTrackDropped, *streamAggrGlobalEnableWindows, dedupInterval, *streamAggrGlobalDropInputLabels, "dedup-global")
-	}
-}
-
-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())
-	}
-	dedupInterval := streamAggrDedupInterval.GetOptionalArg(idx)
-	if dedupInterval > 0 {
-		alias := fmt.Sprintf("dedup-%d", idx+1)
-		var dropLabels []string
-		if streamAggrDropInputLabels.GetOptionalArg(idx) != "" {
-			dropLabels = strings.Split(streamAggrDropInputLabels.GetOptionalArg(idx), "^^")
-		}
-		rwctx.deduplicator = streamaggr.NewDeduplicator(rwctx.pushInternalTrackDropped, *streamAggrGlobalEnableWindows, dedupInterval, dropLabels, 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,
-		DropInputLabels:      *streamAggrGlobalDropInputLabels,
-		IgnoreOldSamples:     *streamAggrGlobalIgnoreOldSamples,
-		IgnoreFirstIntervals: *streamAggrGlobalIgnoreFirstIntervals,
-		KeepInput:            *streamAggrGlobalKeepInput,
-		EnableWindows:        *streamAggrGlobalEnableWindows,
-	}
-
-	sas, err := streamaggr.LoadFromFile(path, pushTimeSeriesToRemoteStoragesTrackDropped, 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))
-	}
-	var dropLabels []string
-	if streamAggrDropInputLabels.GetOptionalArg(idx) != "" {
-		dropLabels = strings.Split(streamAggrDropInputLabels.GetOptionalArg(idx), "^^")
-	}
-	opts := &streamaggr.Options{
-		DedupInterval:        streamAggrDedupInterval.GetOptionalArg(idx),
-		DropInputLabels:      dropLabels,
-		IgnoreOldSamples:     streamAggrIgnoreOldSamples.GetOptionalArg(idx),
-		IgnoreFirstIntervals: streamAggrIgnoreFirstIntervals.GetOptionalArg(idx),
-		KeepInput:            streamAggrKeepInput.GetOptionalArg(idx),
-		EnableWindows:        streamAggrEnableWindows.GetOptionalArg(idx),
-	}
-
-	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 sas, nil
-}

app/vmagent/vmimport/request_handler.go

@@ -8,9 +8,9 @@ import (
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/auth"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/vmimport"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
+	parserCommon "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
+	parser "github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/vmimport"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/vmimport/stream"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/tenantmetrics"
 	"github.com/VictoriaMetrics/metrics"
@@ -26,17 +26,17 @@ var (
 //
 // See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/6
 func InsertHandler(at *auth.Token, req *http.Request) error {
-	extraLabels, err := protoparserutil.GetExtraLabels(req)
+	extraLabels, err := parserCommon.GetExtraLabels(req)
 	if err != nil {
 		return err
 	}
-	encoding := req.Header.Get("Content-Encoding")
-	return stream.Parse(req.Body, encoding, func(rows []vmimport.Row) error {
+	isGzipped := req.Header.Get("Content-Encoding") == "gzip"
+	return stream.Parse(req.Body, isGzipped, func(rows []parser.Row) error {
 		return insertRows(at, rows, extraLabels)
 	})
 }
 
-func insertRows(at *auth.Token, rows []vmimport.Row, extraLabels []prompb.Label) error {
+func insertRows(at *auth.Token, rows []parser.Row, extraLabels []prompbmarshal.Label) error {
 	ctx := common.GetPushCtx()
 	defer common.PutPushCtx(ctx)
 
@@ -50,7 +50,7 @@ func insertRows(at *auth.Token, rows []vmimport.Row, extraLabels []prompb.Label)
 		labelsLen := len(labels)
 		for j := range r.Tags {
 			tag := &r.Tags[j]
-			labels = append(labels, prompb.Label{
+			labels = append(labels, prompbmarshal.Label{
 				Name:  bytesutil.ToUnsafeString(tag.Key),
 				Value: bytesutil.ToUnsafeString(tag.Value),
 			})
@@ -63,12 +63,12 @@ func insertRows(at *auth.Token, rows []vmimport.Row, extraLabels []prompb.Label)
 		}
 		samplesLen := len(samples)
 		for j, value := range values {
-			samples = append(samples, prompb.Sample{
+			samples = append(samples, prompbmarshal.Sample{
 				Value:     value,
 				Timestamp: timestamps[j],
 			})
 		}
-		tssDst = append(tssDst, prompb.TimeSeries{
+		tssDst = append(tssDst, prompbmarshal.TimeSeries{
 			Labels:  labels[labelsLen:],
 			Samples: samples[samplesLen:],
 		})
@@ -76,9 +76,7 @@ func insertRows(at *auth.Token, rows []vmimport.Row, extraLabels []prompb.Label)
 	ctx.WriteRequest.Timeseries = tssDst
 	ctx.Labels = labels
 	ctx.Samples = samples
-	if !remotewrite.TryPush(at, &ctx.WriteRequest) {
-		return remotewrite.ErrQueueFullHTTPRetry
-	}
+	remotewrite.Push(at, &ctx.WriteRequest)
 	rowsInserted.Add(rowsTotal)
 	if at != nil {
 		rowsTenantInserted.Get(at).Add(rowsTotal)

app/vmalert-tool/README.md

@@ -1,3 +0,0 @@
-See vmalert-tool docs [here](https://docs.victoriametrics.com/victoriametrics/vmalert-tool/).
-
-vmalert-tool docs can be edited at [docs/vmalert-tool.md](https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/docs/victoriametrics/vmalert-tool.md).

app/vmalert-tool/deployment/Dockerfile

@@ -1,8 +0,0 @@
-ARG base_image=non-existing
-FROM $base_image
-
-EXPOSE 8880
-
-ENTRYPOINT ["/vmalert-tool-prod"]
-ARG src_binary=non-existing
-COPY $src_binary ./vmalert-tool-prod

app/vmalert-tool/multiarch/Dockerfile

@@ -1,13 +0,0 @@
-# See https://medium.com/on-docker/use-multi-stage-builds-to-inject-ca-certs-ad1e8f01de1b
-ARG certs_image=non-existing
-ARG root_image=non-existing
-FROM $certs_image AS certs
-RUN apk update && apk upgrade && apk --update --no-cache add ca-certificates
-
-FROM $root_image
-COPY --from=certs /etc/ssl/certs/ca-certificates.crt /etc/ssl/certs/ca-certificates.crt
-EXPOSE 8429
-ENTRYPOINT ["/vmalert-tool-prod"]
-ARG TARGETARCH
-ARG BINARY_SUFFIX=non-existing
-COPY vmalert-tool-linux-${TARGETARCH}-prod${BINARY_SUFFIX} ./vmalert-tool-prod

app/vmalert/Makefile

@@ -68,13 +68,12 @@ publish-vmalert:
 
 test-vmalert:
 	go test -v -race -cover ./app/vmalert -loggerLevel=ERROR
-	go test -v -race -cover ./app/vmalert/rule
 	go test -v -race -cover ./app/vmalert/templates
 	go test -v -race -cover ./app/vmalert/datasource
 	go test -v -race -cover ./app/vmalert/notifier
 	go test -v -race -cover ./app/vmalert/config
 	go test -v -race -cover ./app/vmalert/remotewrite
-	go test -v -race -cover ./app/vmalert/vmalertutil
+	go test -v -race -cover ./app/vmalert/utils
 
 run-vmalert: vmalert
 	./bin/vmalert -rule=app/vmalert/config/testdata/rules/rules2-good.rules \
@@ -101,7 +100,8 @@ replay-vmalert: vmalert
 		-remoteWrite.url=http://localhost:8428 \
 		-external.label=cluster=east-1 \
 		-external.label=replica=a \
-		-replay.timeFrom=2024-06-01T00:00:00Z
+		-replay.timeFrom=2021-05-11T07:21:43Z \
+		-replay.timeTo=2021-05-29T18:40:43Z
 
 vmalert-linux-amd64:
 	APP_NAME=vmalert CGO_ENABLED=1 GOOS=linux GOARCH=amd64 $(MAKE) app-local-goos-goarch
@@ -118,9 +118,6 @@ vmalert-linux-ppc64le:
 vmalert-linux-s390x:
 	APP_NAME=vmalert CGO_ENABLED=0 GOOS=linux GOARCH=s390x $(MAKE) app-local-goos-goarch
 
-vmalert-linux-loong64:
-	APP_NAME=vmalert CGO_ENABLED=0 GOOS=linux GOARCH=loong64 $(MAKE) app-local-goos-goarch
-
 vmalert-linux-386:
 	APP_NAME=vmalert CGO_ENABLED=0 GOOS=linux GOARCH=386 $(MAKE) app-local-goos-goarch
 

app/vmalert/alerting.go

@@ -0,0 +1,715 @@
+package main
+
+import (
+	"context"
+	"fmt"
+	"hash/fnv"
+	"sort"
+	"strconv"
+	"strings"
+	"sync"
+	"time"
+
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/config"
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/datasource"
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/notifier"
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/templates"
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/utils"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
+)
+
+// AlertingRule is basic alert entity
+type AlertingRule struct {
+	Type         config.Type
+	RuleID       uint64
+	Name         string
+	Expr         string
+	For          time.Duration
+	Labels       map[string]string
+	Annotations  map[string]string
+	GroupID      uint64
+	GroupName    string
+	EvalInterval time.Duration
+	Debug        bool
+
+	q datasource.Querier
+
+	alertsMu sync.RWMutex
+	// stores list of active alerts
+	alerts map[uint64]*notifier.Alert
+
+	// state stores recent state changes
+	// during evaluations
+	state *ruleState
+
+	metrics *alertingRuleMetrics
+}
+
+type alertingRuleMetrics struct {
+	errors        *utils.Gauge
+	pending       *utils.Gauge
+	active        *utils.Gauge
+	samples       *utils.Gauge
+	seriesFetched *utils.Gauge
+}
+
+func newAlertingRule(qb datasource.QuerierBuilder, group *Group, cfg config.Rule) *AlertingRule {
+	ar := &AlertingRule{
+		Type:         group.Type,
+		RuleID:       cfg.ID,
+		Name:         cfg.Alert,
+		Expr:         cfg.Expr,
+		For:          cfg.For.Duration(),
+		Labels:       cfg.Labels,
+		Annotations:  cfg.Annotations,
+		GroupID:      group.ID(),
+		GroupName:    group.Name,
+		EvalInterval: group.Interval,
+		Debug:        cfg.Debug,
+		q: qb.BuildWithParams(datasource.QuerierParams{
+			DataSourceType:     group.Type.String(),
+			EvaluationInterval: group.Interval,
+			QueryParams:        group.Params,
+			Headers:            group.Headers,
+			Debug:              cfg.Debug,
+		}),
+		alerts:  make(map[uint64]*notifier.Alert),
+		metrics: &alertingRuleMetrics{},
+	}
+
+	if cfg.UpdateEntriesLimit != nil {
+		ar.state = newRuleState(*cfg.UpdateEntriesLimit)
+	} else {
+		ar.state = newRuleState(*ruleUpdateEntriesLimit)
+	}
+
+	labels := fmt.Sprintf(`alertname=%q, group=%q, id="%d"`, ar.Name, group.Name, ar.ID())
+	ar.metrics.pending = utils.GetOrCreateGauge(fmt.Sprintf(`vmalert_alerts_pending{%s}`, labels),
+		func() float64 {
+			ar.alertsMu.RLock()
+			defer ar.alertsMu.RUnlock()
+			var num int
+			for _, a := range ar.alerts {
+				if a.State == notifier.StatePending {
+					num++
+				}
+			}
+			return float64(num)
+		})
+	ar.metrics.active = utils.GetOrCreateGauge(fmt.Sprintf(`vmalert_alerts_firing{%s}`, labels),
+		func() float64 {
+			ar.alertsMu.RLock()
+			defer ar.alertsMu.RUnlock()
+			var num int
+			for _, a := range ar.alerts {
+				if a.State == notifier.StateFiring {
+					num++
+				}
+			}
+			return float64(num)
+		})
+	ar.metrics.errors = utils.GetOrCreateGauge(fmt.Sprintf(`vmalert_alerting_rules_error{%s}`, labels),
+		func() float64 {
+			e := ar.state.getLast()
+			if e.err == nil {
+				return 0
+			}
+			return 1
+		})
+	ar.metrics.samples = utils.GetOrCreateGauge(fmt.Sprintf(`vmalert_alerting_rules_last_evaluation_samples{%s}`, labels),
+		func() float64 {
+			e := ar.state.getLast()
+			return float64(e.samples)
+		})
+	ar.metrics.seriesFetched = utils.GetOrCreateGauge(fmt.Sprintf(`vmalert_alerting_rules_last_evaluation_series_fetched{%s}`, labels),
+		func() float64 {
+			e := ar.state.getLast()
+			if e.seriesFetched == nil {
+				// means seriesFetched is unsupported
+				return -1
+			}
+			seriesFetched := float64(*e.seriesFetched)
+			if seriesFetched == 0 && e.samples > 0 {
+				// `alert: 0.95` will fetch no series
+				// but will get one time series in response.
+				seriesFetched = float64(e.samples)
+			}
+			return seriesFetched
+		})
+	return ar
+}
+
+// Close unregisters rule metrics
+func (ar *AlertingRule) Close() {
+	ar.metrics.active.Unregister()
+	ar.metrics.pending.Unregister()
+	ar.metrics.errors.Unregister()
+	ar.metrics.samples.Unregister()
+	ar.metrics.seriesFetched.Unregister()
+}
+
+// String implements Stringer interface
+func (ar *AlertingRule) String() string {
+	return ar.Name
+}
+
+// ID returns unique Rule ID
+// within the parent Group.
+func (ar *AlertingRule) ID() uint64 {
+	return ar.RuleID
+}
+
+func (ar *AlertingRule) logDebugf(at time.Time, a *notifier.Alert, format string, args ...interface{}) {
+	if !ar.Debug {
+		return
+	}
+	prefix := fmt.Sprintf("DEBUG rule %q:%q (%d) at %v: ",
+		ar.GroupName, ar.Name, ar.RuleID, at.Format(time.RFC3339))
+
+	if a != nil {
+		labelKeys := make([]string, len(a.Labels))
+		var i int
+		for k := range a.Labels {
+			labelKeys[i] = k
+			i++
+		}
+		sort.Strings(labelKeys)
+		labels := make([]string, len(labelKeys))
+		for i, l := range labelKeys {
+			labels[i] = fmt.Sprintf("%s=%q", l, a.Labels[l])
+		}
+		labelsStr := strings.Join(labels, ",")
+		prefix += fmt.Sprintf("alert %d {%s} ", a.ID, labelsStr)
+	}
+	msg := fmt.Sprintf(format, args...)
+	logger.Infof("%s", prefix+msg)
+}
+
+type labelSet struct {
+	// origin labels extracted from received time series
+	// plus extra labels (group labels, service labels like alertNameLabel).
+	// in case of conflicts, origin labels from time series preferred.
+	// used for templating annotations
+	origin map[string]string
+	// processed labels includes origin labels
+	// plus extra labels (group labels, service labels like alertNameLabel).
+	// in case of conflicts, extra labels are preferred.
+	// used as labels attached to notifier.Alert and ALERTS series written to remote storage.
+	processed map[string]string
+}
+
+// toLabels converts labels from given Metric
+// to labelSet which contains original and processed labels.
+func (ar *AlertingRule) toLabels(m datasource.Metric, qFn templates.QueryFn) (*labelSet, error) {
+	ls := &labelSet{
+		origin:    make(map[string]string),
+		processed: make(map[string]string),
+	}
+	for _, l := range m.Labels {
+		ls.origin[l.Name] = l.Value
+		// drop __name__ to be consistent with Prometheus alerting
+		if l.Name == "__name__" {
+			continue
+		}
+		ls.processed[l.Name] = l.Value
+	}
+
+	extraLabels, err := notifier.ExecTemplate(qFn, ar.Labels, notifier.AlertTplData{
+		Labels: ls.origin,
+		Value:  m.Values[0],
+		Expr:   ar.Expr,
+	})
+	if err != nil {
+		return nil, fmt.Errorf("failed to expand labels: %s", err)
+	}
+	for k, v := range extraLabels {
+		ls.processed[k] = v
+		if _, ok := ls.origin[k]; !ok {
+			ls.origin[k] = v
+		}
+	}
+
+	// set additional labels to identify group and rule name
+	if ar.Name != "" {
+		ls.processed[alertNameLabel] = ar.Name
+		if _, ok := ls.origin[alertNameLabel]; !ok {
+			ls.origin[alertNameLabel] = ar.Name
+		}
+	}
+	if !*disableAlertGroupLabel && ar.GroupName != "" {
+		ls.processed[alertGroupNameLabel] = ar.GroupName
+		if _, ok := ls.origin[alertGroupNameLabel]; !ok {
+			ls.origin[alertGroupNameLabel] = ar.GroupName
+		}
+	}
+	return ls, nil
+}
+
+// ExecRange executes alerting rule on the given time range similarly to Exec.
+// It doesn't update internal states of the Rule and meant to be used just
+// to get time series for backfilling.
+// It returns ALERT and ALERT_FOR_STATE time series as result.
+func (ar *AlertingRule) ExecRange(ctx context.Context, start, end time.Time) ([]prompbmarshal.TimeSeries, error) {
+	res, err := ar.q.QueryRange(ctx, ar.Expr, start, end)
+	if err != nil {
+		return nil, err
+	}
+	var result []prompbmarshal.TimeSeries
+	qFn := func(query string) ([]datasource.Metric, error) {
+		return nil, fmt.Errorf("`query` template isn't supported in replay mode")
+	}
+	for _, s := range res.Data {
+		a, err := ar.newAlert(s, nil, time.Time{}, qFn) // initial alert
+		if err != nil {
+			return nil, fmt.Errorf("failed to create alert: %s", err)
+		}
+		if ar.For == 0 { // if alert is instant
+			a.State = notifier.StateFiring
+			for i := range s.Values {
+				result = append(result, ar.alertToTimeSeries(a, s.Timestamps[i])...)
+			}
+			continue
+		}
+
+		// if alert with For > 0
+		prevT := time.Time{}
+		for i := range s.Values {
+			at := time.Unix(s.Timestamps[i], 0)
+			if at.Sub(prevT) > ar.EvalInterval {
+				// reset to Pending if there are gaps > EvalInterval between DPs
+				a.State = notifier.StatePending
+				a.ActiveAt = at
+			} else if at.Sub(a.ActiveAt) >= ar.For {
+				a.State = notifier.StateFiring
+				a.Start = at
+			}
+			prevT = at
+			result = append(result, ar.alertToTimeSeries(a, s.Timestamps[i])...)
+		}
+	}
+	return result, nil
+}
+
+// resolvedRetention is the duration for which a resolved alert instance
+// is kept in memory state and consequently repeatedly sent to the AlertManager.
+const resolvedRetention = 15 * time.Minute
+
+// Exec executes AlertingRule expression via the given Querier.
+// Based on the Querier results AlertingRule maintains notifier.Alerts
+func (ar *AlertingRule) Exec(ctx context.Context, ts time.Time, limit int) ([]prompbmarshal.TimeSeries, error) {
+	start := time.Now()
+	res, req, err := ar.q.Query(ctx, ar.Expr, ts)
+	curState := ruleStateEntry{
+		time:          start,
+		at:            ts,
+		duration:      time.Since(start),
+		samples:       len(res.Data),
+		seriesFetched: res.SeriesFetched,
+		err:           err,
+		curl:          requestToCurl(req),
+	}
+
+	defer func() {
+		ar.state.add(curState)
+	}()
+
+	ar.alertsMu.Lock()
+	defer ar.alertsMu.Unlock()
+
+	if err != nil {
+		return nil, fmt.Errorf("failed to execute query %q: %w", ar.Expr, err)
+	}
+
+	ar.logDebugf(ts, nil, "query returned %d samples (elapsed: %s)", curState.samples, curState.duration)
+
+	for h, a := range ar.alerts {
+		// cleanup inactive alerts from previous Exec
+		if a.State == notifier.StateInactive && ts.Sub(a.ResolvedAt) > resolvedRetention {
+			ar.logDebugf(ts, a, "deleted as inactive")
+			delete(ar.alerts, h)
+		}
+	}
+
+	qFn := func(query string) ([]datasource.Metric, error) {
+		res, _, err := ar.q.Query(ctx, query, ts)
+		return res.Data, err
+	}
+	updated := make(map[uint64]struct{})
+	// update list of active alerts
+	for _, m := range res.Data {
+		ls, err := ar.toLabels(m, qFn)
+		if err != nil {
+			curState.err = fmt.Errorf("failed to expand labels: %s", err)
+			return nil, curState.err
+		}
+		h := hash(ls.processed)
+		if _, ok := updated[h]; ok {
+			// duplicate may be caused by extra labels
+			// conflicting with the metric labels
+			curState.err = fmt.Errorf("labels %v: %w", ls.processed, errDuplicate)
+			return nil, curState.err
+		}
+		updated[h] = struct{}{}
+		if a, ok := ar.alerts[h]; ok {
+			if a.State == notifier.StateInactive {
+				// alert could be in inactive state for resolvedRetention
+				// so when we again receive metrics for it - we switch it
+				// back to notifier.StatePending
+				a.State = notifier.StatePending
+				a.ActiveAt = ts
+				ar.logDebugf(ts, a, "INACTIVE => PENDING")
+			}
+			a.Value = m.Values[0]
+			// re-exec template since Value or query can be used in annotations
+			a.Annotations, err = a.ExecTemplate(qFn, ls.origin, ar.Annotations)
+			if err != nil {
+				return nil, err
+			}
+			continue
+		}
+		a, err := ar.newAlert(m, ls, start, qFn)
+		if err != nil {
+			curState.err = fmt.Errorf("failed to create alert: %w", err)
+			return nil, curState.err
+		}
+		a.ID = h
+		a.State = notifier.StatePending
+		a.ActiveAt = ts
+		ar.alerts[h] = a
+		ar.logDebugf(ts, a, "created in state PENDING")
+	}
+	var numActivePending int
+	for h, a := range ar.alerts {
+		// if alert wasn't updated in this iteration
+		// means it is resolved already
+		if _, ok := updated[h]; !ok {
+			if a.State == notifier.StatePending {
+				// alert was in Pending state - it is not
+				// active anymore
+				delete(ar.alerts, h)
+				ar.logDebugf(ts, a, "PENDING => DELETED: is absent in current evaluation round")
+				continue
+			}
+			if a.State == notifier.StateFiring {
+				a.State = notifier.StateInactive
+				a.ResolvedAt = ts
+				ar.logDebugf(ts, a, "FIRING => INACTIVE: is absent in current evaluation round")
+			}
+			continue
+		}
+		numActivePending++
+		if a.State == notifier.StatePending && ts.Sub(a.ActiveAt) >= ar.For {
+			a.State = notifier.StateFiring
+			a.Start = ts
+			alertsFired.Inc()
+			ar.logDebugf(ts, a, "PENDING => FIRING: %s since becoming active at %v", ts.Sub(a.ActiveAt), a.ActiveAt)
+		}
+	}
+	if limit > 0 && numActivePending > limit {
+		ar.alerts = map[uint64]*notifier.Alert{}
+		curState.err = fmt.Errorf("exec exceeded limit of %d with %d alerts", limit, numActivePending)
+		return nil, curState.err
+	}
+	return ar.toTimeSeries(ts.Unix()), nil
+}
+
+func (ar *AlertingRule) toTimeSeries(timestamp int64) []prompbmarshal.TimeSeries {
+	var tss []prompbmarshal.TimeSeries
+	for _, a := range ar.alerts {
+		if a.State == notifier.StateInactive {
+			continue
+		}
+		ts := ar.alertToTimeSeries(a, timestamp)
+		tss = append(tss, ts...)
+	}
+	return tss
+}
+
+// UpdateWith copies all significant fields.
+// alerts state isn't copied since
+// it should be updated in next 2 Execs
+func (ar *AlertingRule) UpdateWith(r Rule) error {
+	nr, ok := r.(*AlertingRule)
+	if !ok {
+		return fmt.Errorf("BUG: attempt to update alerting rule with wrong type %#v", r)
+	}
+	ar.Expr = nr.Expr
+	ar.For = nr.For
+	ar.Labels = nr.Labels
+	ar.Annotations = nr.Annotations
+	ar.EvalInterval = nr.EvalInterval
+	ar.Debug = nr.Debug
+	ar.q = nr.q
+	ar.state = nr.state
+	return nil
+}
+
+// TODO: consider hashing algorithm in VM
+func hash(labels map[string]string) uint64 {
+	hash := fnv.New64a()
+	keys := make([]string, 0, len(labels))
+	for k := range labels {
+		keys = append(keys, k)
+	}
+	sort.Strings(keys)
+	for _, k := range keys {
+		// drop __name__ to be consistent with Prometheus alerting
+		if k == "__name__" {
+			continue
+		}
+		name, value := k, labels[k]
+		hash.Write([]byte(name))
+		hash.Write([]byte(value))
+		hash.Write([]byte("\xff"))
+	}
+	return hash.Sum64()
+}
+
+func (ar *AlertingRule) newAlert(m datasource.Metric, ls *labelSet, start time.Time, qFn templates.QueryFn) (*notifier.Alert, error) {
+	var err error
+	if ls == nil {
+		ls, err = ar.toLabels(m, qFn)
+		if err != nil {
+			return nil, fmt.Errorf("failed to expand labels: %s", err)
+		}
+	}
+	a := &notifier.Alert{
+		GroupID:  ar.GroupID,
+		Name:     ar.Name,
+		Labels:   ls.processed,
+		Value:    m.Values[0],
+		ActiveAt: start,
+		Expr:     ar.Expr,
+		For:      ar.For,
+	}
+	a.Annotations, err = a.ExecTemplate(qFn, ls.origin, ar.Annotations)
+	return a, err
+}
+
+// AlertAPI generates APIAlert object from alert by its id(hash)
+func (ar *AlertingRule) AlertAPI(id uint64) *APIAlert {
+	ar.alertsMu.RLock()
+	defer ar.alertsMu.RUnlock()
+	a, ok := ar.alerts[id]
+	if !ok {
+		return nil
+	}
+	return ar.newAlertAPI(*a)
+}
+
+// ToAPI returns Rule representation in form of APIRule
+// Isn't thread-safe. Call must be protected by AlertingRule mutex.
+func (ar *AlertingRule) ToAPI() APIRule {
+	lastState := ar.state.getLast()
+	r := APIRule{
+		Type:              "alerting",
+		DatasourceType:    ar.Type.String(),
+		Name:              ar.Name,
+		Query:             ar.Expr,
+		Duration:          ar.For.Seconds(),
+		Labels:            ar.Labels,
+		Annotations:       ar.Annotations,
+		LastEvaluation:    lastState.time,
+		EvaluationTime:    lastState.duration.Seconds(),
+		Health:            "ok",
+		State:             "inactive",
+		Alerts:            ar.AlertsToAPI(),
+		LastSamples:       lastState.samples,
+		LastSeriesFetched: lastState.seriesFetched,
+		MaxUpdates:        ar.state.size(),
+		Updates:           ar.state.getAll(),
+		Debug:             ar.Debug,
+
+		// encode as strings to avoid rounding in JSON
+		ID:      fmt.Sprintf("%d", ar.ID()),
+		GroupID: fmt.Sprintf("%d", ar.GroupID),
+	}
+	if lastState.err != nil {
+		r.LastError = lastState.err.Error()
+		r.Health = "err"
+	}
+	// satisfy APIRule.State logic
+	if len(r.Alerts) > 0 {
+		r.State = notifier.StatePending.String()
+		stateFiring := notifier.StateFiring.String()
+		for _, a := range r.Alerts {
+			if a.State == stateFiring {
+				r.State = stateFiring
+				break
+			}
+		}
+	}
+	return r
+}
+
+// AlertsToAPI generates list of APIAlert objects from existing alerts
+func (ar *AlertingRule) AlertsToAPI() []*APIAlert {
+	var alerts []*APIAlert
+	ar.alertsMu.RLock()
+	for _, a := range ar.alerts {
+		if a.State == notifier.StateInactive {
+			continue
+		}
+		alerts = append(alerts, ar.newAlertAPI(*a))
+	}
+	ar.alertsMu.RUnlock()
+	return alerts
+}
+
+func (ar *AlertingRule) newAlertAPI(a notifier.Alert) *APIAlert {
+	aa := &APIAlert{
+		// encode as strings to avoid rounding
+		ID:      fmt.Sprintf("%d", a.ID),
+		GroupID: fmt.Sprintf("%d", a.GroupID),
+		RuleID:  fmt.Sprintf("%d", ar.RuleID),
+
+		Name:        a.Name,
+		Expression:  ar.Expr,
+		Labels:      a.Labels,
+		Annotations: a.Annotations,
+		State:       a.State.String(),
+		ActiveAt:    a.ActiveAt,
+		Restored:    a.Restored,
+		Value:       strconv.FormatFloat(a.Value, 'f', -1, 32),
+	}
+	if alertURLGeneratorFn != nil {
+		aa.SourceLink = alertURLGeneratorFn(a)
+	}
+	return aa
+}
+
+const (
+	// alertMetricName is the metric name for synthetic alert timeseries.
+	alertMetricName = "ALERTS"
+	// alertForStateMetricName is the metric name for 'for' state of alert.
+	alertForStateMetricName = "ALERTS_FOR_STATE"
+
+	// alertNameLabel is the label name indicating the name of an alert.
+	alertNameLabel = "alertname"
+	// alertStateLabel is the label name indicating the state of an alert.
+	alertStateLabel = "alertstate"
+
+	// alertGroupNameLabel defines the label name attached for generated time series.
+	// attaching this label may be disabled via `-disableAlertgroupLabel` flag.
+	alertGroupNameLabel = "alertgroup"
+)
+
+// alertToTimeSeries converts the given alert with the given timestamp to time series
+func (ar *AlertingRule) alertToTimeSeries(a *notifier.Alert, timestamp int64) []prompbmarshal.TimeSeries {
+	var tss []prompbmarshal.TimeSeries
+	tss = append(tss, alertToTimeSeries(a, timestamp))
+	if ar.For > 0 {
+		tss = append(tss, alertForToTimeSeries(a, timestamp))
+	}
+	return tss
+}
+
+func alertToTimeSeries(a *notifier.Alert, timestamp int64) prompbmarshal.TimeSeries {
+	labels := make(map[string]string)
+	for k, v := range a.Labels {
+		labels[k] = v
+	}
+	labels["__name__"] = alertMetricName
+	labels[alertStateLabel] = a.State.String()
+	return newTimeSeries([]float64{1}, []int64{timestamp}, labels)
+}
+
+// alertForToTimeSeries returns a timeseries that represents
+// state of active alerts, where value is time when alert become active
+func alertForToTimeSeries(a *notifier.Alert, timestamp int64) prompbmarshal.TimeSeries {
+	labels := make(map[string]string)
+	for k, v := range a.Labels {
+		labels[k] = v
+	}
+	labels["__name__"] = alertForStateMetricName
+	return newTimeSeries([]float64{float64(a.ActiveAt.Unix())}, []int64{timestamp}, labels)
+}
+
+// Restore restores the value of ActiveAt field for active alerts,
+// based on previously written time series `alertForStateMetricName`.
+// Only rules with For > 0 can be restored.
+func (ar *AlertingRule) Restore(ctx context.Context, q datasource.Querier, ts time.Time, lookback time.Duration) error {
+	if ar.For < 1 {
+		return nil
+	}
+
+	ar.alertsMu.Lock()
+	defer ar.alertsMu.Unlock()
+
+	if len(ar.alerts) < 1 {
+		return nil
+	}
+
+	for _, a := range ar.alerts {
+		if a.Restored || a.State != notifier.StatePending {
+			continue
+		}
+
+		var labelsFilter []string
+		for k, v := range a.Labels {
+			labelsFilter = append(labelsFilter, fmt.Sprintf("%s=%q", k, v))
+		}
+		sort.Strings(labelsFilter)
+		expr := fmt.Sprintf("last_over_time(%s{%s}[%ds])",
+			alertForStateMetricName, strings.Join(labelsFilter, ","), int(lookback.Seconds()))
+
+		ar.logDebugf(ts, nil, "restoring alert state via query %q", expr)
+
+		res, _, err := q.Query(ctx, expr, ts)
+		if err != nil {
+			return err
+		}
+
+		qMetrics := res.Data
+		if len(qMetrics) < 1 {
+			ar.logDebugf(ts, nil, "no response was received from restore query")
+			continue
+		}
+
+		// only one series expected in response
+		m := qMetrics[0]
+		// __name__ supposed to be alertForStateMetricName
+		m.DelLabel("__name__")
+
+		// we assume that restore query contains all label matchers,
+		// so all received labels will match anyway if their number is equal.
+		if len(m.Labels) != len(a.Labels) {
+			ar.logDebugf(ts, nil, "state restore query returned not expected label-set %v", m.Labels)
+			continue
+		}
+		a.ActiveAt = time.Unix(int64(m.Values[0]), 0)
+		a.Restored = true
+		logger.Infof("alert %q (%d) restored to state at %v", a.Name, a.ID, a.ActiveAt)
+	}
+	return nil
+}
+
+// alertsToSend walks through the current alerts of AlertingRule
+// and returns only those which should be sent to notifier.
+// Isn't concurrent safe.
+func (ar *AlertingRule) alertsToSend(ts time.Time, resolveDuration, resendDelay time.Duration) []notifier.Alert {
+	needsSending := func(a *notifier.Alert) bool {
+		if a.State == notifier.StatePending {
+			return false
+		}
+		if a.ResolvedAt.After(a.LastSent) {
+			return true
+		}
+		return a.LastSent.Add(resendDelay).Before(ts)
+	}
+
+	var alerts []notifier.Alert
+	for _, a := range ar.alerts {
+		if !needsSending(a) {
+			continue
+		}
+		a.End = ts.Add(resolveDuration)
+		if a.State == notifier.StateInactive {
+			a.End = a.ResolvedAt
+		}
+		a.LastSent = ts
+		alerts = append(alerts, *a)
+	}
+	return alerts
+}

app/vmalert/alerting_test.go

@@ -0,0 +1,980 @@
+package main
+
+import (
+	"context"
+	"errors"
+	"reflect"
+	"sort"
+	"strings"
+	"sync"
+	"testing"
+	"time"
+
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/config"
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/datasource"
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/notifier"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promutils"
+)
+
+func TestAlertingRule_ToTimeSeries(t *testing.T) {
+	timestamp := time.Now()
+	testCases := []struct {
+		rule  *AlertingRule
+		alert *notifier.Alert
+		expTS []prompbmarshal.TimeSeries
+	}{
+		{
+			newTestAlertingRule("instant", 0),
+			&notifier.Alert{State: notifier.StateFiring},
+			[]prompbmarshal.TimeSeries{
+				newTimeSeries([]float64{1}, []int64{timestamp.UnixNano()}, map[string]string{
+					"__name__":      alertMetricName,
+					alertStateLabel: notifier.StateFiring.String(),
+				}),
+			},
+		},
+		{
+			newTestAlertingRule("instant extra labels", 0),
+			&notifier.Alert{State: notifier.StateFiring, Labels: map[string]string{
+				"job":      "foo",
+				"instance": "bar",
+			}},
+			[]prompbmarshal.TimeSeries{
+				newTimeSeries([]float64{1}, []int64{timestamp.UnixNano()}, map[string]string{
+					"__name__":      alertMetricName,
+					alertStateLabel: notifier.StateFiring.String(),
+					"job":           "foo",
+					"instance":      "bar",
+				}),
+			},
+		},
+		{
+			newTestAlertingRule("instant labels override", 0),
+			&notifier.Alert{State: notifier.StateFiring, Labels: map[string]string{
+				alertStateLabel: "foo",
+				"__name__":      "bar",
+			}},
+			[]prompbmarshal.TimeSeries{
+				newTimeSeries([]float64{1}, []int64{timestamp.UnixNano()}, map[string]string{
+					"__name__":      alertMetricName,
+					alertStateLabel: notifier.StateFiring.String(),
+				}),
+			},
+		},
+		{
+			newTestAlertingRule("for", time.Second),
+			&notifier.Alert{State: notifier.StateFiring, ActiveAt: timestamp.Add(time.Second)},
+			[]prompbmarshal.TimeSeries{
+				newTimeSeries([]float64{1}, []int64{timestamp.UnixNano()}, map[string]string{
+					"__name__":      alertMetricName,
+					alertStateLabel: notifier.StateFiring.String(),
+				}),
+				newTimeSeries([]float64{float64(timestamp.Add(time.Second).Unix())},
+					[]int64{timestamp.UnixNano()},
+					map[string]string{
+						"__name__": alertForStateMetricName,
+					}),
+			},
+		},
+		{
+			newTestAlertingRule("for pending", 10*time.Second),
+			&notifier.Alert{State: notifier.StatePending, ActiveAt: timestamp.Add(time.Second)},
+			[]prompbmarshal.TimeSeries{
+				newTimeSeries([]float64{1}, []int64{timestamp.UnixNano()}, map[string]string{
+					"__name__":      alertMetricName,
+					alertStateLabel: notifier.StatePending.String(),
+				}),
+				newTimeSeries([]float64{float64(timestamp.Add(time.Second).Unix())},
+					[]int64{timestamp.UnixNano()},
+					map[string]string{
+						"__name__": alertForStateMetricName,
+					}),
+			},
+		},
+	}
+	for _, tc := range testCases {
+		t.Run(tc.rule.Name, func(t *testing.T) {
+			tc.rule.alerts[tc.alert.ID] = tc.alert
+			tss := tc.rule.toTimeSeries(timestamp.Unix())
+			if err := compareTimeSeries(t, tc.expTS, tss); err != nil {
+				t.Fatalf("timeseries missmatch: %s", err)
+			}
+		})
+	}
+}
+
+func TestAlertingRule_Exec(t *testing.T) {
+	const defaultStep = 5 * time.Millisecond
+	type testAlert struct {
+		labels []string
+		alert  *notifier.Alert
+	}
+	testCases := []struct {
+		rule      *AlertingRule
+		steps     [][]datasource.Metric
+		expAlerts []testAlert
+	}{
+		{
+			newTestAlertingRule("empty", 0),
+			[][]datasource.Metric{},
+			nil,
+		},
+		{
+			newTestAlertingRule("empty labels", 0),
+			[][]datasource.Metric{
+				{datasource.Metric{Values: []float64{1}, Timestamps: []int64{1}}},
+			},
+			[]testAlert{
+				{alert: &notifier.Alert{State: notifier.StateFiring}},
+			},
+		},
+		{
+			newTestAlertingRule("single-firing", 0),
+			[][]datasource.Metric{
+				{metricWithLabels(t, "name", "foo")},
+			},
+			[]testAlert{
+				{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StateFiring}},
+			},
+		},
+		{
+			newTestAlertingRule("single-firing=>inactive", 0),
+			[][]datasource.Metric{
+				{metricWithLabels(t, "name", "foo")},
+				{},
+			},
+			[]testAlert{
+				{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StateInactive}},
+			},
+		},
+		{
+			newTestAlertingRule("single-firing=>inactive=>firing", 0),
+			[][]datasource.Metric{
+				{metricWithLabels(t, "name", "foo")},
+				{},
+				{metricWithLabels(t, "name", "foo")},
+			},
+			[]testAlert{
+				{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StateFiring}},
+			},
+		},
+		{
+			newTestAlertingRule("single-firing=>inactive=>firing=>inactive", 0),
+			[][]datasource.Metric{
+				{metricWithLabels(t, "name", "foo")},
+				{},
+				{metricWithLabels(t, "name", "foo")},
+				{},
+			},
+			[]testAlert{
+				{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StateInactive}},
+			},
+		},
+		{
+			newTestAlertingRule("single-firing=>inactive=>firing=>inactive=>inactive", 0),
+			[][]datasource.Metric{
+				{metricWithLabels(t, "name", "foo")},
+				{},
+				{metricWithLabels(t, "name", "foo")},
+				{},
+				{},
+			},
+			[]testAlert{
+				{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StateInactive}},
+			},
+		},
+		{
+			newTestAlertingRule("single-firing=>inactive=>firing=>inactive=>empty=>firing", 0),
+			[][]datasource.Metric{
+				{metricWithLabels(t, "name", "foo")},
+				{},
+				{metricWithLabels(t, "name", "foo")},
+				{},
+				{},
+				{metricWithLabels(t, "name", "foo")},
+			},
+			[]testAlert{
+				{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StateFiring}},
+			},
+		},
+		{
+			newTestAlertingRule("multiple-firing", 0),
+			[][]datasource.Metric{
+				{
+					metricWithLabels(t, "name", "foo"),
+					metricWithLabels(t, "name", "foo1"),
+					metricWithLabels(t, "name", "foo2"),
+				},
+			},
+			[]testAlert{
+				{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StateFiring}},
+				{labels: []string{"name", "foo1"}, alert: &notifier.Alert{State: notifier.StateFiring}},
+				{labels: []string{"name", "foo2"}, alert: &notifier.Alert{State: notifier.StateFiring}},
+			},
+		},
+		{
+			newTestAlertingRule("multiple-steps-firing", 0),
+			[][]datasource.Metric{
+				{metricWithLabels(t, "name", "foo")},
+				{metricWithLabels(t, "name", "foo1")},
+				{metricWithLabels(t, "name", "foo2")},
+			},
+			// 1: fire first alert
+			// 2: fire second alert, set first inactive
+			// 3: fire third alert, set second inactive
+			[]testAlert{
+				{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StateInactive}},
+				{labels: []string{"name", "foo1"}, alert: &notifier.Alert{State: notifier.StateInactive}},
+				{labels: []string{"name", "foo2"}, alert: &notifier.Alert{State: notifier.StateFiring}},
+			},
+		},
+		{
+			newTestAlertingRule("for-pending", time.Minute),
+			[][]datasource.Metric{
+				{metricWithLabels(t, "name", "foo")},
+			},
+			[]testAlert{
+				{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StatePending}},
+			},
+		},
+		{
+			newTestAlertingRule("for-fired", defaultStep),
+			[][]datasource.Metric{
+				{metricWithLabels(t, "name", "foo")},
+				{metricWithLabels(t, "name", "foo")},
+			},
+			[]testAlert{
+				{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StateFiring}},
+			},
+		},
+		{
+			newTestAlertingRule("for-pending=>empty", time.Second),
+			[][]datasource.Metric{
+				{metricWithLabels(t, "name", "foo")},
+				{metricWithLabels(t, "name", "foo")},
+				// empty step to reset and delete pending alerts
+				{},
+			},
+			nil,
+		},
+		{
+			newTestAlertingRule("for-pending=>firing=>inactive", defaultStep),
+			[][]datasource.Metric{
+				{metricWithLabels(t, "name", "foo")},
+				{metricWithLabels(t, "name", "foo")},
+				// empty step to reset pending alerts
+				{},
+			},
+			[]testAlert{
+				{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StateInactive}},
+			},
+		},
+		{
+			newTestAlertingRule("for-pending=>firing=>inactive=>pending", defaultStep),
+			[][]datasource.Metric{
+				{metricWithLabels(t, "name", "foo")},
+				{metricWithLabels(t, "name", "foo")},
+				// empty step to reset pending alerts
+				{},
+				{metricWithLabels(t, "name", "foo")},
+			},
+			[]testAlert{
+				{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StatePending}},
+			},
+		},
+		{
+			newTestAlertingRule("for-pending=>firing=>inactive=>pending=>firing", defaultStep),
+			[][]datasource.Metric{
+				{metricWithLabels(t, "name", "foo")},
+				{metricWithLabels(t, "name", "foo")},
+				// empty step to reset pending alerts
+				{},
+				{metricWithLabels(t, "name", "foo")},
+				{metricWithLabels(t, "name", "foo")},
+			},
+			[]testAlert{
+				{labels: []string{"name", "foo"}, alert: &notifier.Alert{State: notifier.StateFiring}},
+			},
+		},
+	}
+	fakeGroup := Group{Name: "TestRule_Exec"}
+	for _, tc := range testCases {
+		t.Run(tc.rule.Name, func(t *testing.T) {
+			fq := &fakeQuerier{}
+			tc.rule.q = fq
+			tc.rule.GroupID = fakeGroup.ID()
+			for _, step := range tc.steps {
+				fq.reset()
+				fq.add(step...)
+				if _, err := tc.rule.Exec(context.TODO(), time.Now(), 0); err != nil {
+					t.Fatalf("unexpected err: %s", err)
+				}
+				// artificial delay between applying steps
+				time.Sleep(defaultStep)
+			}
+			if len(tc.rule.alerts) != len(tc.expAlerts) {
+				t.Fatalf("expected %d alerts; got %d", len(tc.expAlerts), len(tc.rule.alerts))
+			}
+			expAlerts := make(map[uint64]*notifier.Alert)
+			for _, ta := range tc.expAlerts {
+				labels := make(map[string]string)
+				for i := 0; i < len(ta.labels); i += 2 {
+					k, v := ta.labels[i], ta.labels[i+1]
+					labels[k] = v
+				}
+				labels[alertNameLabel] = tc.rule.Name
+				h := hash(labels)
+				expAlerts[h] = ta.alert
+			}
+			for key, exp := range expAlerts {
+				got, ok := tc.rule.alerts[key]
+				if !ok {
+					t.Fatalf("expected to have key %d", key)
+				}
+				if got.State != exp.State {
+					t.Fatalf("expected state %d; got %d", exp.State, got.State)
+				}
+			}
+		})
+	}
+}
+
+func TestAlertingRule_ExecRange(t *testing.T) {
+	testCases := []struct {
+		rule      *AlertingRule
+		data      []datasource.Metric
+		expAlerts []*notifier.Alert
+	}{
+		{
+			newTestAlertingRule("empty", 0),
+			[]datasource.Metric{},
+			nil,
+		},
+		{
+			newTestAlertingRule("empty labels", 0),
+			[]datasource.Metric{
+				{Values: []float64{1}, Timestamps: []int64{1}},
+			},
+			[]*notifier.Alert{
+				{State: notifier.StateFiring},
+			},
+		},
+		{
+			newTestAlertingRule("single-firing", 0),
+			[]datasource.Metric{
+				metricWithLabels(t, "name", "foo"),
+			},
+			[]*notifier.Alert{
+				{
+					Labels: map[string]string{"name": "foo"},
+					State:  notifier.StateFiring,
+				},
+			},
+		},
+		{
+			newTestAlertingRule("single-firing-on-range", 0),
+			[]datasource.Metric{
+				{Values: []float64{1, 1, 1}, Timestamps: []int64{1e3, 2e3, 3e3}},
+			},
+			[]*notifier.Alert{
+				{State: notifier.StateFiring},
+				{State: notifier.StateFiring},
+				{State: notifier.StateFiring},
+			},
+		},
+		{
+			newTestAlertingRule("for-pending", time.Second),
+			[]datasource.Metric{
+				{Values: []float64{1, 1, 1}, Timestamps: []int64{1, 3, 5}},
+			},
+			[]*notifier.Alert{
+				{State: notifier.StatePending, ActiveAt: time.Unix(1, 0)},
+				{State: notifier.StatePending, ActiveAt: time.Unix(3, 0)},
+				{State: notifier.StatePending, ActiveAt: time.Unix(5, 0)},
+			},
+		},
+		{
+			newTestAlertingRule("for-firing", 3*time.Second),
+			[]datasource.Metric{
+				{Values: []float64{1, 1, 1}, Timestamps: []int64{1, 3, 5}},
+			},
+			[]*notifier.Alert{
+				{State: notifier.StatePending, ActiveAt: time.Unix(1, 0)},
+				{State: notifier.StatePending, ActiveAt: time.Unix(1, 0)},
+				{State: notifier.StateFiring, ActiveAt: time.Unix(1, 0)},
+			},
+		},
+		{
+			newTestAlertingRule("for=>pending=>firing=>pending=>firing=>pending", time.Second),
+			[]datasource.Metric{
+				{Values: []float64{1, 1, 1, 1, 1}, Timestamps: []int64{1, 2, 5, 6, 20}},
+			},
+			[]*notifier.Alert{
+				{State: notifier.StatePending, ActiveAt: time.Unix(1, 0)},
+				{State: notifier.StateFiring, ActiveAt: time.Unix(1, 0)},
+				{State: notifier.StatePending, ActiveAt: time.Unix(5, 0)},
+				{State: notifier.StateFiring, ActiveAt: time.Unix(5, 0)},
+				{State: notifier.StatePending, ActiveAt: time.Unix(20, 0)},
+			},
+		},
+		{
+			newTestAlertingRule("multi-series-for=>pending=>pending=>firing", 3*time.Second),
+			[]datasource.Metric{
+				{Values: []float64{1, 1, 1}, Timestamps: []int64{1, 3, 5}},
+				{Values: []float64{1, 1}, Timestamps: []int64{1, 5},
+					Labels: []datasource.Label{{Name: "foo", Value: "bar"}},
+				},
+			},
+			[]*notifier.Alert{
+				{State: notifier.StatePending, ActiveAt: time.Unix(1, 0)},
+				{State: notifier.StatePending, ActiveAt: time.Unix(1, 0)},
+				{State: notifier.StateFiring, ActiveAt: time.Unix(1, 0)},
+				//
+				{State: notifier.StatePending, ActiveAt: time.Unix(1, 0),
+					Labels: map[string]string{
+						"foo": "bar",
+					}},
+				{State: notifier.StatePending, ActiveAt: time.Unix(5, 0),
+					Labels: map[string]string{
+						"foo": "bar",
+					}},
+			},
+		},
+		{
+			newTestRuleWithLabels("multi-series-firing", "source", "vm"),
+			[]datasource.Metric{
+				{Values: []float64{1, 1}, Timestamps: []int64{1, 100}},
+				{Values: []float64{1, 1}, Timestamps: []int64{1, 5},
+					Labels: []datasource.Label{{Name: "foo", Value: "bar"}},
+				},
+			},
+			[]*notifier.Alert{
+				{State: notifier.StateFiring, Labels: map[string]string{
+					"source": "vm",
+				}},
+				{State: notifier.StateFiring, Labels: map[string]string{
+					"source": "vm",
+				}},
+				//
+				{State: notifier.StateFiring, Labels: map[string]string{
+					"foo":    "bar",
+					"source": "vm",
+				}},
+				{State: notifier.StateFiring, Labels: map[string]string{
+					"foo":    "bar",
+					"source": "vm",
+				}},
+			},
+		},
+	}
+	fakeGroup := Group{Name: "TestRule_ExecRange"}
+	for _, tc := range testCases {
+		t.Run(tc.rule.Name, func(t *testing.T) {
+			fq := &fakeQuerier{}
+			tc.rule.q = fq
+			tc.rule.GroupID = fakeGroup.ID()
+			fq.add(tc.data...)
+			gotTS, err := tc.rule.ExecRange(context.TODO(), time.Now(), time.Now())
+			if err != nil {
+				t.Fatalf("unexpected err: %s", err)
+			}
+			var expTS []prompbmarshal.TimeSeries
+			var j int
+			for _, series := range tc.data {
+				for _, timestamp := range series.Timestamps {
+					a := tc.expAlerts[j]
+					if a.Labels == nil {
+						a.Labels = make(map[string]string)
+					}
+					a.Labels[alertNameLabel] = tc.rule.Name
+					expTS = append(expTS, tc.rule.alertToTimeSeries(a, timestamp)...)
+					j++
+				}
+			}
+			if len(gotTS) != len(expTS) {
+				t.Fatalf("expected %d time series; got %d", len(expTS), len(gotTS))
+			}
+			for i := range expTS {
+				got, exp := gotTS[i], expTS[i]
+				if !reflect.DeepEqual(got, exp) {
+					t.Fatalf("%d: expected \n%v but got \n%v", i, exp, got)
+				}
+			}
+		})
+	}
+}
+
+func TestGroup_Restore(t *testing.T) {
+	defaultTS := time.Now()
+	fqr := &fakeQuerierWithRegistry{}
+	fn := func(rules []config.Rule, expAlerts map[uint64]*notifier.Alert) {
+		t.Helper()
+		defer fqr.reset()
+
+		for _, r := range rules {
+			fqr.set(r.Expr, metricWithValueAndLabels(t, 0, "__name__", r.Alert))
+		}
+
+		fg := newGroup(config.Group{Name: "TestRestore", Rules: rules}, fqr, time.Second, nil)
+		wg := sync.WaitGroup{}
+		wg.Add(1)
+		go func() {
+			nts := func() []notifier.Notifier { return []notifier.Notifier{&fakeNotifier{}} }
+			fg.start(context.Background(), nts, nil, fqr)
+			wg.Done()
+		}()
+		fg.close()
+		wg.Wait()
+
+		gotAlerts := make(map[uint64]*notifier.Alert)
+		for _, rs := range fg.Rules {
+			alerts := rs.(*AlertingRule).alerts
+			for k, v := range alerts {
+				if !v.Restored {
+					// set not restored alerts to predictable timestamp
+					v.ActiveAt = defaultTS
+				}
+				gotAlerts[k] = v
+			}
+		}
+
+		if len(gotAlerts) != len(expAlerts) {
+			t.Fatalf("expected %d alerts; got %d", len(expAlerts), len(gotAlerts))
+		}
+		for key, exp := range expAlerts {
+			got, ok := gotAlerts[key]
+			if !ok {
+				t.Fatalf("expected to have key %d", key)
+			}
+			if got.State != notifier.StatePending {
+				t.Fatalf("expected state %d; got %d", notifier.StatePending, got.State)
+			}
+			if got.ActiveAt != exp.ActiveAt {
+				t.Fatalf("expected ActiveAt %v; got %v", exp.ActiveAt, got.ActiveAt)
+			}
+		}
+	}
+
+	stateMetric := func(name string, value time.Time, labels ...string) datasource.Metric {
+		labels = append(labels, "__name__", alertForStateMetricName)
+		labels = append(labels, alertNameLabel, name)
+		labels = append(labels, alertGroupNameLabel, "TestRestore")
+		return metricWithValueAndLabels(t, float64(value.Unix()), labels...)
+	}
+
+	// one active alert, no previous state
+	fn(
+		[]config.Rule{{Alert: "foo", Expr: "foo", For: promutils.NewDuration(time.Second)}},
+		map[uint64]*notifier.Alert{
+			hash(map[string]string{alertNameLabel: "foo", alertGroupNameLabel: "TestRestore"}): {
+				ActiveAt: defaultTS,
+			},
+		})
+	fqr.reset()
+
+	// one active alert with state restore
+	ts := time.Now().Truncate(time.Hour)
+	fqr.set(`last_over_time(ALERTS_FOR_STATE{alertgroup="TestRestore",alertname="foo"}[3600s])`,
+		stateMetric("foo", ts))
+	fn(
+		[]config.Rule{{Alert: "foo", Expr: "foo", For: promutils.NewDuration(time.Second)}},
+		map[uint64]*notifier.Alert{
+			hash(map[string]string{alertNameLabel: "foo", alertGroupNameLabel: "TestRestore"}): {
+				ActiveAt: ts},
+		})
+
+	// two rules, two active alerts, one with state restored
+	ts = time.Now().Truncate(time.Hour)
+	fqr.set(`last_over_time(ALERTS_FOR_STATE{alertgroup="TestRestore",alertname="bar"}[3600s])`,
+		stateMetric("foo", ts))
+	fn(
+		[]config.Rule{
+			{Alert: "foo", Expr: "foo", For: promutils.NewDuration(time.Second)},
+			{Alert: "bar", Expr: "bar", For: promutils.NewDuration(time.Second)},
+		},
+		map[uint64]*notifier.Alert{
+			hash(map[string]string{alertNameLabel: "foo", alertGroupNameLabel: "TestRestore"}): {
+				ActiveAt: defaultTS,
+			},
+			hash(map[string]string{alertNameLabel: "bar", alertGroupNameLabel: "TestRestore"}): {
+				ActiveAt: ts},
+		})
+
+	// two rules, two active alerts, two with state restored
+	ts = time.Now().Truncate(time.Hour)
+	fqr.set(`last_over_time(ALERTS_FOR_STATE{alertgroup="TestRestore",alertname="foo"}[3600s])`,
+		stateMetric("foo", ts))
+	fqr.set(`last_over_time(ALERTS_FOR_STATE{alertgroup="TestRestore",alertname="bar"}[3600s])`,
+		stateMetric("bar", ts))
+	fn(
+		[]config.Rule{
+			{Alert: "foo", Expr: "foo", For: promutils.NewDuration(time.Second)},
+			{Alert: "bar", Expr: "bar", For: promutils.NewDuration(time.Second)},
+		},
+		map[uint64]*notifier.Alert{
+			hash(map[string]string{alertNameLabel: "foo", alertGroupNameLabel: "TestRestore"}): {
+				ActiveAt: ts,
+			},
+			hash(map[string]string{alertNameLabel: "bar", alertGroupNameLabel: "TestRestore"}): {
+				ActiveAt: ts},
+		})
+
+	// one active alert but wrong state restore
+	ts = time.Now().Truncate(time.Hour)
+	fqr.set(`last_over_time(ALERTS_FOR_STATE{alertname="bar",alertgroup="TestRestore"}[3600s])`,
+		stateMetric("wrong alert", ts))
+	fn(
+		[]config.Rule{{Alert: "foo", Expr: "foo", For: promutils.NewDuration(time.Second)}},
+		map[uint64]*notifier.Alert{
+			hash(map[string]string{alertNameLabel: "foo", alertGroupNameLabel: "TestRestore"}): {
+				ActiveAt: defaultTS,
+			},
+		})
+
+	// one active alert with labels
+	ts = time.Now().Truncate(time.Hour)
+	fqr.set(`last_over_time(ALERTS_FOR_STATE{alertgroup="TestRestore",alertname="foo",env="dev"}[3600s])`,
+		stateMetric("foo", ts, "env", "dev"))
+	fn(
+		[]config.Rule{{Alert: "foo", Expr: "foo", Labels: map[string]string{"env": "dev"}, For: promutils.NewDuration(time.Second)}},
+		map[uint64]*notifier.Alert{
+			hash(map[string]string{alertNameLabel: "foo", alertGroupNameLabel: "TestRestore", "env": "dev"}): {
+				ActiveAt: ts,
+			},
+		})
+
+	// one active alert with restore labels missmatch
+	ts = time.Now().Truncate(time.Hour)
+	fqr.set(`last_over_time(ALERTS_FOR_STATE{alertgroup="TestRestore",alertname="foo",env="dev"}[3600s])`,
+		stateMetric("foo", ts, "env", "dev", "team", "foo"))
+	fn(
+		[]config.Rule{{Alert: "foo", Expr: "foo", Labels: map[string]string{"env": "dev"}, For: promutils.NewDuration(time.Second)}},
+		map[uint64]*notifier.Alert{
+			hash(map[string]string{alertNameLabel: "foo", alertGroupNameLabel: "TestRestore", "env": "dev"}): {
+				ActiveAt: defaultTS,
+			},
+		})
+}
+
+func TestAlertingRule_Exec_Negative(t *testing.T) {
+	fq := &fakeQuerier{}
+	ar := newTestAlertingRule("test", 0)
+	ar.Labels = map[string]string{"job": "test"}
+	ar.q = fq
+
+	// successful attempt
+	fq.add(metricWithValueAndLabels(t, 1, "__name__", "foo", "job", "bar"))
+	_, err := ar.Exec(context.TODO(), time.Now(), 0)
+	if err != nil {
+		t.Fatal(err)
+	}
+
+	// label `job` will collide with rule extra label and will make both time series equal
+	fq.add(metricWithValueAndLabels(t, 1, "__name__", "foo", "job", "baz"))
+	_, err = ar.Exec(context.TODO(), time.Now(), 0)
+	if !errors.Is(err, errDuplicate) {
+		t.Fatalf("expected to have %s error; got %s", errDuplicate, err)
+	}
+
+	fq.reset()
+
+	expErr := "connection reset by peer"
+	fq.setErr(errors.New(expErr))
+	_, err = ar.Exec(context.TODO(), time.Now(), 0)
+	if err == nil {
+		t.Fatalf("expected to get err; got nil")
+	}
+	if !strings.Contains(err.Error(), expErr) {
+		t.Fatalf("expected to get err %q; got %q insterad", expErr, err)
+	}
+}
+
+func TestAlertingRuleLimit(t *testing.T) {
+	fq := &fakeQuerier{}
+	ar := newTestAlertingRule("test", 0)
+	ar.Labels = map[string]string{"job": "test"}
+	ar.q = fq
+	ar.For = time.Minute
+	testCases := []struct {
+		limit  int
+		err    string
+		tssNum int
+	}{
+		{
+			limit:  0,
+			tssNum: 4,
+		},
+		{
+			limit:  -1,
+			tssNum: 4,
+		},
+		{
+			limit:  1,
+			err:    "exec exceeded limit of 1 with 2 alerts",
+			tssNum: 0,
+		},
+		{
+			limit:  4,
+			tssNum: 4,
+		},
+	}
+	var (
+		err       error
+		timestamp = time.Now()
+	)
+	fq.add(metricWithValueAndLabels(t, 1, "__name__", "foo", "job", "bar"))
+	fq.add(metricWithValueAndLabels(t, 1, "__name__", "foo", "bar", "job"))
+	for _, testCase := range testCases {
+		_, err = ar.Exec(context.TODO(), timestamp, testCase.limit)
+		if err != nil && !strings.EqualFold(err.Error(), testCase.err) {
+			t.Fatal(err)
+		}
+	}
+	fq.reset()
+}
+
+func TestAlertingRule_Template(t *testing.T) {
+	testCases := []struct {
+		rule      *AlertingRule
+		metrics   []datasource.Metric
+		expAlerts map[uint64]*notifier.Alert
+	}{
+		{
+			&AlertingRule{
+				Name: "common",
+				Labels: map[string]string{
+					"region": "east",
+				},
+				Annotations: map[string]string{
+					"summary": `{{ $labels.alertname }}: Too high connection number for "{{ $labels.instance }}"`,
+				},
+				alerts: make(map[uint64]*notifier.Alert),
+			},
+			[]datasource.Metric{
+				metricWithValueAndLabels(t, 1, "instance", "foo"),
+				metricWithValueAndLabels(t, 1, "instance", "bar"),
+			},
+			map[uint64]*notifier.Alert{
+				hash(map[string]string{alertNameLabel: "common", "region": "east", "instance": "foo"}): {
+					Annotations: map[string]string{
+						"summary": `common: Too high connection number for "foo"`,
+					},
+					Labels: map[string]string{
+						alertNameLabel: "common",
+						"region":       "east",
+						"instance":     "foo",
+					},
+				},
+				hash(map[string]string{alertNameLabel: "common", "region": "east", "instance": "bar"}): {
+					Annotations: map[string]string{
+						"summary": `common: Too high connection number for "bar"`,
+					},
+					Labels: map[string]string{
+						alertNameLabel: "common",
+						"region":       "east",
+						"instance":     "bar",
+					},
+				},
+			},
+		},
+		{
+			&AlertingRule{
+				Name: "override label",
+				Labels: map[string]string{
+					"instance": "{{ $labels.instance }}",
+				},
+				Annotations: map[string]string{
+					"summary":     `{{ $labels.__name__ }}: Too high connection number for "{{ $labels.instance }}"`,
+					"description": `{{ $labels.alertname}}: It is {{ $value }} connections for "{{ $labels.instance }}"`,
+				},
+				alerts: make(map[uint64]*notifier.Alert),
+			},
+			[]datasource.Metric{
+				metricWithValueAndLabels(t, 2, "__name__", "first", "instance", "foo", alertNameLabel, "override"),
+				metricWithValueAndLabels(t, 10, "__name__", "second", "instance", "bar", alertNameLabel, "override"),
+			},
+			map[uint64]*notifier.Alert{
+				hash(map[string]string{alertNameLabel: "override label", "instance": "foo"}): {
+					Labels: map[string]string{
+						alertNameLabel: "override label",
+						"instance":     "foo",
+					},
+					Annotations: map[string]string{
+						"summary":     `first: Too high connection number for "foo"`,
+						"description": `override: It is 2 connections for "foo"`,
+					},
+				},
+				hash(map[string]string{alertNameLabel: "override label", "instance": "bar"}): {
+					Labels: map[string]string{
+						alertNameLabel: "override label",
+						"instance":     "bar",
+					},
+					Annotations: map[string]string{
+						"summary":     `second: Too high connection number for "bar"`,
+						"description": `override: It is 10 connections for "bar"`,
+					},
+				},
+			},
+		},
+		{
+			&AlertingRule{
+				Name:      "OriginLabels",
+				GroupName: "Testing",
+				Labels: map[string]string{
+					"instance": "{{ $labels.instance }}",
+				},
+				Annotations: map[string]string{
+					"summary": `Alert "{{ $labels.alertname }}({{ $labels.alertgroup }})" for instance {{ $labels.instance }}`,
+				},
+				alerts: make(map[uint64]*notifier.Alert),
+			},
+			[]datasource.Metric{
+				metricWithValueAndLabels(t, 1,
+					alertNameLabel, "originAlertname",
+					alertGroupNameLabel, "originGroupname",
+					"instance", "foo"),
+			},
+			map[uint64]*notifier.Alert{
+				hash(map[string]string{
+					alertNameLabel:      "OriginLabels",
+					alertGroupNameLabel: "Testing",
+					"instance":          "foo"}): {
+					Labels: map[string]string{
+						alertNameLabel:      "OriginLabels",
+						alertGroupNameLabel: "Testing",
+						"instance":          "foo",
+					},
+					Annotations: map[string]string{
+						"summary": `Alert "originAlertname(originGroupname)" for instance foo`,
+					},
+				},
+			},
+		},
+	}
+	fakeGroup := Group{Name: "TestRule_Exec"}
+	for _, tc := range testCases {
+		t.Run(tc.rule.Name, func(t *testing.T) {
+			fq := &fakeQuerier{}
+			tc.rule.GroupID = fakeGroup.ID()
+			tc.rule.q = fq
+			tc.rule.state = newRuleState(10)
+			fq.add(tc.metrics...)
+			if _, err := tc.rule.Exec(context.TODO(), time.Now(), 0); err != nil {
+				t.Fatalf("unexpected err: %s", err)
+			}
+			for hash, expAlert := range tc.expAlerts {
+				gotAlert := tc.rule.alerts[hash]
+				if gotAlert == nil {
+					t.Fatalf("alert %d is missing; labels: %v; annotations: %v",
+						hash, expAlert.Labels, expAlert.Annotations)
+				}
+				if !reflect.DeepEqual(expAlert.Annotations, gotAlert.Annotations) {
+					t.Fatalf("expected to have annotations %#v; got %#v", expAlert.Annotations, gotAlert.Annotations)
+				}
+				if !reflect.DeepEqual(expAlert.Labels, gotAlert.Labels) {
+					t.Fatalf("expected to have labels %#v; got %#v", expAlert.Labels, gotAlert.Labels)
+				}
+			}
+		})
+	}
+}
+
+func TestAlertsToSend(t *testing.T) {
+	ts := time.Now()
+	f := func(alerts, expAlerts []*notifier.Alert, resolveDuration, resendDelay time.Duration) {
+		t.Helper()
+		ar := &AlertingRule{alerts: make(map[uint64]*notifier.Alert)}
+		for i, a := range alerts {
+			ar.alerts[uint64(i)] = a
+		}
+		gotAlerts := ar.alertsToSend(ts, resolveDuration, resendDelay)
+		if gotAlerts == nil && expAlerts == nil {
+			return
+		}
+		if len(gotAlerts) != len(expAlerts) {
+			t.Fatalf("expected to get %d alerts; got %d instead",
+				len(expAlerts), len(gotAlerts))
+		}
+		sort.Slice(expAlerts, func(i, j int) bool {
+			return expAlerts[i].Name < expAlerts[j].Name
+		})
+		sort.Slice(gotAlerts, func(i, j int) bool {
+			return gotAlerts[i].Name < gotAlerts[j].Name
+		})
+		for i, exp := range expAlerts {
+			got := gotAlerts[i]
+			if got.LastSent != exp.LastSent {
+				t.Fatalf("expected LastSent to be %v; got %v", exp.LastSent, got.LastSent)
+			}
+			if got.End != exp.End {
+				t.Fatalf("expected End to be %v; got %v", exp.End, got.End)
+			}
+		}
+	}
+
+	f( // send firing alert with custom resolve time
+		[]*notifier.Alert{{State: notifier.StateFiring}},
+		[]*notifier.Alert{{LastSent: ts, End: ts.Add(5 * time.Minute)}},
+		5*time.Minute, time.Minute,
+	)
+	f( // resolve inactive alert at the current timestamp
+		[]*notifier.Alert{{State: notifier.StateInactive, ResolvedAt: ts}},
+		[]*notifier.Alert{{LastSent: ts, End: ts}},
+		time.Minute, time.Minute,
+	)
+	f( // mixed case of firing and resolved alerts. Names are added for deterministic sorting
+		[]*notifier.Alert{{Name: "a", State: notifier.StateFiring}, {Name: "b", State: notifier.StateInactive, ResolvedAt: ts}},
+		[]*notifier.Alert{{Name: "a", LastSent: ts, End: ts.Add(5 * time.Minute)}, {Name: "b", LastSent: ts, End: ts}},
+		5*time.Minute, time.Minute,
+	)
+	f( // mixed case of pending and resolved alerts. Names are added for deterministic sorting
+		[]*notifier.Alert{{Name: "a", State: notifier.StatePending}, {Name: "b", State: notifier.StateInactive, ResolvedAt: ts}},
+		[]*notifier.Alert{{Name: "b", LastSent: ts, End: ts}},
+		5*time.Minute, time.Minute,
+	)
+	f( // attempt to send alert that was already sent in the resendDelay interval
+		[]*notifier.Alert{{State: notifier.StateFiring, LastSent: ts.Add(-time.Second)}},
+		nil,
+		time.Minute, time.Minute,
+	)
+	f( // attempt to send alert that was sent out of the resendDelay interval
+		[]*notifier.Alert{{State: notifier.StateFiring, LastSent: ts.Add(-2 * time.Minute)}},
+		[]*notifier.Alert{{LastSent: ts, End: ts.Add(time.Minute)}},
+		time.Minute, time.Minute,
+	)
+	f( // alert must be sent even if resendDelay interval is 0
+		[]*notifier.Alert{{State: notifier.StateFiring, LastSent: ts.Add(-time.Second)}},
+		[]*notifier.Alert{{LastSent: ts, End: ts.Add(time.Minute)}},
+		time.Minute, 0,
+	)
+	f( // inactive alert which has been sent already
+		[]*notifier.Alert{{State: notifier.StateInactive, LastSent: ts.Add(-time.Second), ResolvedAt: ts.Add(-2 * time.Second)}},
+		nil,
+		time.Minute, time.Minute,
+	)
+	f( // inactive alert which has been resolved after last send
+		[]*notifier.Alert{{State: notifier.StateInactive, LastSent: ts.Add(-time.Second), ResolvedAt: ts}},
+		[]*notifier.Alert{{LastSent: ts, End: ts}},
+		time.Minute, time.Minute,
+	)
+}
+
+func newTestRuleWithLabels(name string, labels ...string) *AlertingRule {
+	r := newTestAlertingRule(name, 0)
+	r.Labels = make(map[string]string)
+	for i := 0; i < len(labels); i += 2 {
+		r.Labels[labels[i]] = labels[i+1]
+	}
+	return r
+}
+
+func newTestAlertingRule(name string, waitFor time.Duration) *AlertingRule {
+	return &AlertingRule{
+		Name:         name,
+		For:          waitFor,
+		EvalInterval: waitFor,
+		alerts:       make(map[uint64]*notifier.Alert),
+		state:        newRuleState(10),
+	}
+}

app/vmalert/config/config.go

@@ -1,11 +1,9 @@
 package config
 
 import (
-	"bytes"
-	"flag"
+	"crypto/md5"
 	"fmt"
 	"hash/fnv"
-	"io"
 	"net/url"
 	"sort"
 	"strings"
@@ -13,27 +11,21 @@ import (
 	"gopkg.in/yaml.v2"
 
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/config/log"
-	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/vmalertutil"
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/utils"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/envtemplate"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promutil"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promutils"
 )
 
-var defaultRuleType = flag.String("rule.defaultRuleType", "prometheus", `Default type for rule expressions, can be overridden via "type" parameter on the group level, see https://docs.victoriametrics.com/victoriametrics/vmalert/#groups. Supported values: "graphite", "prometheus" and "vlogs".`)
-
 // Group contains list of Rules grouped into
 // entity with one name and evaluation interval
 type Group struct {
-	Type       Type `yaml:"type,omitempty"`
-	File       string
-	Name       string             `yaml:"name"`
-	Interval   *promutil.Duration `yaml:"interval,omitempty"`
-	EvalOffset *promutil.Duration `yaml:"eval_offset,omitempty"`
-	// EvalDelay will adjust the `time` parameter of rule evaluation requests to compensate intentional query delay from datasource.
-	// see https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5155
-	EvalDelay   *promutil.Duration `yaml:"eval_delay,omitempty"`
-	Limit       int                `yaml:"limit,omitempty"`
-	Rules       []Rule             `yaml:"rules"`
-	Concurrency int                `yaml:"concurrency"`
+	Type        Type `yaml:"type,omitempty"`
+	File        string
+	Name        string              `yaml:"name"`
+	Interval    *promutils.Duration `yaml:"interval,omitempty"`
+	Limit       int                 `yaml:"limit,omitempty"`
+	Rules       []Rule              `yaml:"rules"`
+	Concurrency int                 `yaml:"concurrency"`
 	// Labels is a set of label value pairs, that will be added to every rule.
 	// It has priority over the external labels.
 	Labels map[string]string `yaml:"labels"`
@@ -46,16 +38,12 @@ type Group struct {
 	Headers []Header `yaml:"headers,omitempty"`
 	// NotifierHeaders contains optional HTTP headers sent to notifiers for generated notifications
 	NotifierHeaders []Header `yaml:"notifier_headers,omitempty"`
-	// EvalAlignment will make the timestamp of group query requests be aligned with interval
-	EvalAlignment *bool `yaml:"eval_alignment,omitempty"`
-	// Debug enables debug logs for the group
-	Debug bool `yaml:"debug,omitempty"`
 	// Catches all undefined fields and must be empty after parsing.
-	XXX map[string]any `yaml:",inline"`
+	XXX map[string]interface{} `yaml:",inline"`
 }
 
 // UnmarshalYAML implements the yaml.Unmarshaler interface.
-func (g *Group) UnmarshalYAML(unmarshal func(any) error) error {
+func (g *Group) UnmarshalYAML(unmarshal func(interface{}) error) error {
 	type group Group
 	if err := unmarshal((*group)(g)); err != nil {
 		return err
@@ -64,39 +52,22 @@ func (g *Group) UnmarshalYAML(unmarshal func(any) error) error {
 	if err != nil {
 		return fmt.Errorf("failed to marshal group configuration for checksum: %w", err)
 	}
+	// change default value to prometheus datasource.
 	if g.Type.Get() == "" {
-		g.Type = NewRawType(*defaultRuleType)
+		g.Type.Set(NewPrometheusType())
 	}
-	h := fnv.New64a()
+
+	h := md5.New()
 	h.Write(b)
 	g.Checksum = fmt.Sprintf("%x", h.Sum(nil))
 	return nil
 }
 
-// Validate checks configuration errors for group and internal rules
+// Validate check for internal Group or Rule configuration errors
 func (g *Group) Validate(validateTplFn ValidateTplFn, validateExpressions bool) error {
 	if g.Name == "" {
 		return fmt.Errorf("group name must be set")
 	}
-	if g.Interval.Duration() < 0 {
-		return fmt.Errorf("interval shouldn't be lower than 0")
-	}
-	if g.EvalOffset.Duration() < 0 {
-		return fmt.Errorf("eval_offset shouldn't be lower than 0")
-	}
-	// if `eval_offset` is set, interval won't use global evaluationInterval flag and must bigger than offset.
-	if g.EvalOffset.Duration() > g.Interval.Duration() {
-		return fmt.Errorf("eval_offset should be smaller than interval; now eval_offset: %v, interval: %v", g.EvalOffset.Duration(), g.Interval.Duration())
-	}
-	if g.EvalOffset != nil && g.EvalDelay != nil {
-		return fmt.Errorf("eval_offset cannot be used with eval_delay")
-	}
-	if g.Limit < 0 {
-		return fmt.Errorf("invalid limit %d, shouldn't be less than 0", g.Limit)
-	}
-	if g.Concurrency < 0 {
-		return fmt.Errorf("invalid concurrency %d, shouldn't be less than 0", g.Concurrency)
-	}
 
 	uniqueRules := map[uint64]struct{}{}
 	for _, r := range g.Rules {
@@ -105,26 +76,26 @@ func (g *Group) Validate(validateTplFn ValidateTplFn, validateExpressions bool)
 			ruleName = r.Alert
 		}
 		if _, ok := uniqueRules[r.ID]; ok {
-			return fmt.Errorf("%q is a duplicate in group", r.String())
+			return fmt.Errorf("%q is a duplicate within the group %q", r.String(), g.Name)
 		}
 		uniqueRules[r.ID] = struct{}{}
 		if err := r.Validate(); err != nil {
-			return fmt.Errorf("invalid rule %q: %w", ruleName, err)
+			return fmt.Errorf("invalid rule %q.%q: %w", g.Name, ruleName, err)
 		}
 		if validateExpressions {
 			// its needed only for tests.
 			// because correct types must be inherited after unmarshalling.
 			exprValidator := g.Type.ValidateExpr
 			if err := exprValidator(r.Expr); err != nil {
-				return fmt.Errorf("invalid expression for rule  %q: %w", ruleName, err)
+				return fmt.Errorf("invalid expression for rule %q.%q: %w", g.Name, ruleName, err)
 			}
 		}
 		if validateTplFn != nil {
 			if err := validateTplFn(r.Annotations); err != nil {
-				return fmt.Errorf("invalid annotations for rule  %q: %w", ruleName, err)
+				return fmt.Errorf("invalid annotations for rule %q.%q: %w", g.Name, ruleName, err)
 			}
 			if err := validateTplFn(r.Labels); err != nil {
-				return fmt.Errorf("invalid labels for rule  %q: %w", ruleName, err)
+				return fmt.Errorf("invalid labels for rule %q.%q: %w", g.Name, ruleName, err)
 			}
 		}
 	}
@@ -134,26 +105,24 @@ func (g *Group) Validate(validateTplFn ValidateTplFn, validateExpressions bool)
 // Rule describes entity that represent either
 // recording rule or alerting rule.
 type Rule struct {
-	ID     uint64
-	Record string             `yaml:"record,omitempty"`
-	Alert  string             `yaml:"alert,omitempty"`
-	Expr   string             `yaml:"expr"`
-	For    *promutil.Duration `yaml:"for,omitempty"`
-	// Alert will continue firing for this long even when the alerting expression no longer has results.
-	KeepFiringFor *promutil.Duration `yaml:"keep_firing_for,omitempty"`
-	Labels        map[string]string  `yaml:"labels,omitempty"`
-	Annotations   map[string]string  `yaml:"annotations,omitempty"`
-	Debug         *bool              `yaml:"debug,omitempty"`
+	ID          uint64
+	Record      string              `yaml:"record,omitempty"`
+	Alert       string              `yaml:"alert,omitempty"`
+	Expr        string              `yaml:"expr"`
+	For         *promutils.Duration `yaml:"for,omitempty"`
+	Labels      map[string]string   `yaml:"labels,omitempty"`
+	Annotations map[string]string   `yaml:"annotations,omitempty"`
+	Debug       bool                `yaml:"debug,omitempty"`
 	// UpdateEntriesLimit defines max number of rule's state updates stored in memory.
 	// Overrides `-rule.updateEntriesLimit`.
 	UpdateEntriesLimit *int `yaml:"update_entries_limit,omitempty"`
 
 	// Catches all undefined fields and must be empty after parsing.
-	XXX map[string]any `yaml:",inline"`
+	XXX map[string]interface{} `yaml:",inline"`
 }
 
 // UnmarshalYAML implements the yaml.Unmarshaler interface.
-func (r *Rule) UnmarshalYAML(unmarshal func(any) error) error {
+func (r *Rule) UnmarshalYAML(unmarshal func(interface{}) error) error {
 	type rule Rule
 	if err := unmarshal((*rule)(r)); err != nil {
 		return err
@@ -241,22 +210,22 @@ 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)
+		return nil, fmt.Errorf("failed to read from the config: %s", err)
 	}
 	return parse(files, validateTplFn, validateExpressions)
 }
 
 // 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)
+		return nil, fmt.Errorf("failed to read from the config: %s", err)
 	}
 	groups, err := parse(files, validateTplFn, validateExpressions)
 	if err != nil {
-		return nil, fmt.Errorf("failed to parse %s: %w", pathPatterns, err)
+		return nil, fmt.Errorf("failed to parse %s: %s", pathPatterns, err)
 	}
 	if len(groups) < 1 {
 		cLogger.Warnf("no groups found in %s", strings.Join(pathPatterns, ";"))
@@ -265,7 +234,7 @@ func Parse(pathPatterns []string, validateTplFn ValidateTplFn, validateExpressio
 }
 
 func parse(files map[string][]byte, validateTplFn ValidateTplFn, validateExpressions bool) ([]Group, error) {
-	errGroup := new(vmalertutil.ErrGroup)
+	errGroup := new(utils.ErrGroup)
 	var groups []Group
 	for file, data := range files {
 		uniqueGroups := map[string]struct{}{}
@@ -291,38 +260,33 @@ func parse(files map[string][]byte, validateTplFn ValidateTplFn, validateExpress
 	if err := errGroup.Err(); err != nil {
 		return nil, err
 	}
+	sort.SliceStable(groups, func(i, j int) bool {
+		if groups[i].File != groups[j].File {
+			return groups[i].File < groups[j].File
+		}
+		return groups[i].Name < groups[j].Name
+	})
 	return groups, nil
 }
 
 func parseConfig(data []byte) ([]Group, error) {
-	data = envtemplate.ReplaceBytes(data)
-
-	var result []Group
-	type cfgFile struct {
+	data, err := envtemplate.ReplaceBytes(data)
+	if err != nil {
+		return nil, fmt.Errorf("cannot expand environment vars: %w", err)
+	}
+	g := struct {
 		Groups []Group `yaml:"groups"`
 		// Catches all undefined fields and must be empty after parsing.
-		XXX map[string]any `yaml:",inline"`
+		XXX map[string]interface{} `yaml:",inline"`
+	}{}
+	err = yaml.Unmarshal(data, &g)
+	if err != nil {
+		return nil, err
 	}
-
-	decoder := yaml.NewDecoder(bytes.NewReader(data))
-	for {
-		var cf cfgFile
-		if err := decoder.Decode(&cf); err != nil {
-			if err == io.EOF { // EOF indicates no more documents to read
-				break
-			}
-			return nil, err
-		}
-		if err := checkOverflow(cf.XXX, "config"); err != nil {
-			return nil, err
-		}
-		result = append(result, cf.Groups...)
-	}
-
-	return result, nil
+	return g.Groups, checkOverflow(g.XXX, "config")
 }
 
-func checkOverflow(m map[string]any, ctx string) error {
+func checkOverflow(m map[string]interface{}, ctx string) error {
 	if len(m) > 0 {
 		var keys []string
 		for k := range m {

app/vmalert/config/config_test.go

@@ -9,19 +9,26 @@ import (
 	"testing"
 	"time"
 
+	"gopkg.in/yaml.v2"
+
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/notifier"
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/templates"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promutil"
-	"gopkg.in/yaml.v2"
+	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promutils"
 )
 
 func TestMain(m *testing.M) {
-	if err := templates.Load([]string{"testdata/templates/*good.tmpl"}, url.URL{}); err != nil {
+	if err := templates.Load([]string{"testdata/templates/*good.tmpl"}, true); err != nil {
 		os.Exit(1)
 	}
 	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) {
@@ -39,34 +46,6 @@ groups:
 		w.Write([]byte(`
 groups:
   - name: TestGroup
-    rules:
-      - record: conns
-        expr: max(vm_tcplistener_conns)`))
-	})
-	mux.HandleFunc("/good-multi-doc", func(w http.ResponseWriter, _ *http.Request) {
-		w.Write([]byte(`
-groups:
-  - name: foo
-    rules:
-      - record: conns
-        expr: max(vm_tcplistener_conns)
----
-groups:
-  - name: bar
-    rules:
-      - record: conns
-        expr: max(vm_tcplistener_conns)`))
-	})
-	mux.HandleFunc("/bad-multi-doc", func(w http.ResponseWriter, _ *http.Request) {
-		w.Write([]byte(`
-bad_field:
-  - name: foo
-    rules:
-      - record: conns
-        expr: max(vm_tcplistener_conns)
----
-groups:
-  - name: bar
     rules:
       - record: conns
         expr: max(vm_tcplistener_conns)`))
@@ -75,421 +54,406 @@ groups:
 	srv := httptest.NewServer(mux)
 	defer srv.Close()
 
-	f := func(urls []string, expErr bool) {
-		for i, u := range urls {
-			urls[i] = srv.URL + u
-		}
-		_, err := Parse(urls, notifier.ValidateTemplates, true)
-		if err != nil && !expErr {
-			t.Fatalf("error parsing URLs %s", err)
-		}
-		if err == nil && expErr {
-			t.Fatalf("expecting error parsing URLs but got none")
-		}
+	if _, err := Parse([]string{srv.URL + "/good-alert", srv.URL + "/good-rr"}, notifier.ValidateTemplates, true); err != nil {
+		t.Errorf("error parsing URLs %s", err)
 	}
 
-	f([]string{"/good-alert", "/good-rr", "/good-multi-doc"}, false)
-	f([]string{"/bad"}, true)
-	f([]string{"/bad-multi-doc"}, true)
-	f([]string{"/good-alert", "/bad"}, true)
-}
-
-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)
+	if _, err := Parse([]string{srv.URL + "/bad"}, notifier.ValidateTemplates, true); err == nil {
+		t.Errorf("expected parsing error: %s", err)
 	}
 }
 
-func TestParse_Failure(t *testing.T) {
-	f := func(paths []string, errStrExpected string) {
-		t.Helper()
-
-		_, err := Parse(paths, notifier.ValidateTemplates, true)
+func TestParseBad(t *testing.T) {
+	testCases := []struct {
+		path   []string
+		expErr string
+	}{
+		{
+			[]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 read",
+		},
+	}
+	for _, tc := range testCases {
+		_, err := Parse(tc.path, notifier.ValidateTemplates, true)
 		if err == nil {
-			t.Fatalf("expected to get error")
+			t.Errorf("expected to get error")
+			return
 		}
-		if !strings.Contains(err.Error(), errStrExpected) {
-			t.Fatalf("expected err to contain %q; got %q instead", errStrExpected, err)
+		if !strings.Contains(err.Error(), tc.expErr) {
+			t.Errorf("expected err to contain %q; got %q instead", tc.expErr, 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/rules/vlog-rules0-bad.rules"}, "bad LogsQL expr")
-	f([]string{"testdata/dir/rules6-bad.rules"}, "missing ':' in header")
-	f([]string{"testdata/rules/rules-multi-doc-bad.rules"}, "unknown fields")
-	f([]string{"testdata/rules/rules-multi-doc-duplicates-bad.rules"}, "duplicate")
-	f([]string{"http://unreachable-url"}, "failed to")
 }
 
-func TestRuleValidate(t *testing.T) {
+func TestRule_Validate(t *testing.T) {
 	if err := (&Rule{}).Validate(); err == nil {
-		t.Fatalf("expected empty name error")
+		t.Errorf("expected empty name error")
 	}
 	if err := (&Rule{Alert: "alert"}).Validate(); err == nil {
-		t.Fatalf("expected empty expr error")
+		t.Errorf("expected empty expr error")
 	}
 	if err := (&Rule{Alert: "alert", Expr: "test>0"}).Validate(); err != nil {
-		t.Fatalf("expected valid rule; got %s", err)
+		t.Errorf("expected valid rule; got %s", err)
 	}
 }
 
-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: "both record and alert are not set",
-		Rules: []Rule{
-			{
-				Expr: "sum(up == 0 ) by (host)",
-				For:  promutil.NewDuration(10 * time.Millisecond),
-			},
-			{
-				Expr: "sumSeries(time('foo.bar',10))",
-			},
+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",
 		},
-	}, false, "invalid rule")
-
-	f(&Group{
-		Name:     "negative interval",
-		Interval: promutil.NewDuration(-1),
-	}, false, "interval shouldn't be lower than 0")
-
-	f(&Group{
-		Name:       "wrong eval_offset",
-		Interval:   promutil.NewDuration(time.Minute),
-		EvalOffset: promutil.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",
-			},
-			{
-				Alert: "alert",
-				Expr:  "up == 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 thanos",
-		Type: NewRawType("thanos"),
-		Rules: []Rule{
-			{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-				"description": "{{ value|query }}",
-			}},
-		},
-	}, true, "unknown datasource type")
-
-	// validate expressions
-	f(&Group{
-		Name: "test prometheus expr",
-		Type: NewPrometheusType(),
-		Rules: []Rule{
-			{
-				Record: "record",
-				Expr:   "up | 0",
-			},
-		},
-	}, true, "bad prometheus expr")
-
-	f(&Group{
-		Name: "test graphite expr",
-		Type: NewGraphiteType(),
-		Rules: []Rule{
-			{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-				"description": "some-description",
-			}},
-		},
-	}, true, "bad graphite expr")
-
-	f(&Group{
-		Name: "test vlogs expr",
-		Type: NewVLogsType(),
-		Rules: []Rule{
-			{Alert: "alert", Expr: "stats count(*) as requests"},
-		},
-	}, true, "bad LogsQL expr")
-
-	f(&Group{
-		Name: "test vlogs expr",
-		Type: NewVLogsType(),
-		Rules: []Rule{
-			{Alert: "alert", Expr: "_time: 1m | stats by (path, _time: 1m) count(*) as requests"},
-		},
-	}, true, "bad LogsQL expr")
-
-	f(&Group{
-		Name: "test graphite with prometheus expr",
-		Type: NewGraphiteType(),
-		Rules: []Rule{
-			{
-				Record: "r1",
-				ID:     1,
-				Expr:   "sumSeries(time('foo.bar',10))",
-				For:    promutil.NewDuration(10 * time.Millisecond),
-			},
-			{
-				Record: "r2",
-				ID:     2,
-				Expr:   "sum(up == 0 ) by (host)",
-			},
-		},
-	}, true, "bad graphite expr")
-
-	f(&Group{
-		Name: "test vlogs with prometheus exp",
-		Type: NewVLogsType(),
-		Rules: []Rule{
-			{
-				Record: "r1",
-				Expr:   "sum(up == 0 ) by (host)",
-				For:    promutil.NewDuration(10 * time.Millisecond),
-			},
-		},
-	}, true, "bad LogsQL expr")
-
-	f(&Group{
-		Name: "test prometheus with vlogs exp",
-		Type: NewPrometheusType(),
-		Rules: []Rule{
-			{
-				Record: "r1",
-				Expr:   "* | stats by (path) count()",
-				For:    promutil.NewDuration(10 * time.Millisecond),
-			},
-		},
-	}, true, "bad prometheus 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 }}",
+		{
+			group: &Group{
+				Name: "test",
+				Rules: []Rule{
+					{
+						Record: "record",
+						Expr:   "up | 0",
+					},
 				},
 			},
+			expErr: "",
 		},
-	}, false, false)
-
-	// validate annotations
-	f(&Group{
-		Name: "test",
-		Rules: []Rule{
-			{
-				Alert: "alert",
-				Expr:  "up == 1",
-				Labels: map[string]string{
-					"summary": `
+		{
+			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": `
 {{ 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)",
+					},
 				},
 			},
 		},
-	}, true, false)
-
-	// validate expressions
-	f(&Group{
-		Name: "test prometheus",
-		Type: NewPrometheusType(),
-		Rules: []Rule{
-			{Alert: "alert", Expr: "up == 1", Labels: map[string]string{
-				"description": "{{ value|query }}",
-			}},
+		{
+			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",
 		},
-	}, false, true)
-	f(&Group{
-		Name: "test victorialogs",
-		Type: NewVLogsType(),
-		Rules: []Rule{
-			{Alert: "alert", Expr: " _time: 1m | stats count(*) as requests", Labels: map[string]string{
-				"description": "{{ value|query }}",
-			}},
-		},
-	}, false, true)
-}
-
-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)
+	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)
 		}
 	}
+}
 
-	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: promutil.NewDuration(time.Minute), KeepFiringFor: promutil.NewDuration(time.Minute),
-	}, Rule{Alert: "alert", Expr: "up == 1"})
+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)},
+			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 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/fsurl/url.go

@@ -49,7 +49,7 @@ func (fs *FS) Read(files []string) (map[string][]byte, error) {
 				path, resp.StatusCode, http.StatusOK, data)
 		}
 		if err != nil {
-			return nil, fmt.Errorf("cannot read %q: %w", path, err)
+			return nil, fmt.Errorf("cannot read %q: %s", path, err)
 		}
 		result[path] = data
 	}

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 ...any) {
+func (l *Logger) Errorf(format string, args ...interface{}) {
 	if l.isDisabled() {
 		return
 	}
@@ -37,7 +37,7 @@ func (l *Logger) Errorf(format string, args ...any) {
 }
 
 // Warnf logs warning message.
-func (l *Logger) Warnf(format string, args ...any) {
+func (l *Logger) Warnf(format string, args ...interface{}) {
 	if l.isDisabled() {
 		return
 	}
@@ -45,7 +45,7 @@ func (l *Logger) Warnf(format string, args ...any) {
 }
 
 // Infof logs info message.
-func (l *Logger) Infof(format string, args ...any) {
+func (l *Logger) Infof(format string, args ...interface{}) {
 	if l.isDisabled() {
 		return
 	}
@@ -54,6 +54,6 @@ func (l *Logger) Infof(format string, args ...any) {
 
 // Panicf logs panic message and panics.
 // Panicf can't be suppressed
-func (l *Logger) Panicf(format string, args ...any) {
+func (l *Logger) Panicf(format string, args ...interface{}) {
 	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.Fatalf("expected output to be empty; got %q", testOutput.String())
+				t.Errorf("expected output to be empty; got %q", testOutput.String())
+				return
 			}
 		}
 		if !strings.Contains(testOutput.String(), exp) {
-			t.Fatalf("output %q should contain %q", testOutput.String(), exp)
+			t.Errorf("output %q should contain %q", testOutput.String(), exp)
 		}
 		fmt.Println(testOutput.String())
 		testOutput.Reset()

app/vmalert/config/testdata/dir/rules0-bad.rules

@@ -7,9 +7,9 @@ groups:
         labels:
           label: bar
         annotations:
-          summary: "{{ }}"
+          summary: "{{ $value }"
           description: "{{$labels}}"
-      - alert: UnknownAnnotationsFunction
+      - alert: UnkownAnnotationsFunction
         for: 5m
         expr: vm_rows > 0
         labels:

app/vmalert/config/testdata/dir/rules2-bad.rules

@@ -1,7 +1,7 @@
 groups:
   - name: group
     rules:
-      - alert: UnknownLabelFunction
+      - alert: UnkownLabelFunction
         for: 5m
         expr: vm_rows > 0
         labels:

app/vmalert/config/testdata/rules/kube-good.rules

@@ -1158,9 +1158,9 @@
           $labels.pod }}.'
         runbook_url: https://github.com/kubernetes-monitoring/kubernetes-mixin/tree/master/runbook.md#alert-name-cputhrottlinghigh
       expr: |
-        sum(increase(container_cpu_cfs_throttled_periods_total{container!="", }[5m])) by (cluster, container, pod, namespace)
+        sum(increase(container_cpu_cfs_throttled_periods_total{container!="", }[5m])) by (container, pod, namespace)
           /
-        sum(increase(container_cpu_cfs_periods_total{}[5m])) by (cluster, container, pod, namespace)
+        sum(increase(container_cpu_cfs_periods_total{}[5m])) by (container, pod, namespace)
           > ( 25 / 100 )
       for: 15m
       labels:

app/vmalert/config/testdata/rules/rules-multi-doc-bad.rules

@@ -1,29 +0,0 @@
-groups:
-  - name: groupTest
-    rules:
-      - alert: VMRows
-        for: 1ms
-        expr: vm_rows > 0
-        labels:
-          label: bar
-          host: "{{ $labels.instance }}"
-        annotations:
-          summary: "{{ $value }}"
-invalid-field-1: invalid-value-1
-invalid-field-2: invalid-value-2
----
-groups:
-  - name: TestGroup
-    interval: 2s
-    concurrency: 2
-    type: graphite
-    rules:
-      - alert: Conns
-        expr: filterSeries(sumSeries(host.receiver.interface.cons),'last','>', 500)
-        for: 3m
-
-        annotations:
-          summary: Too high connection number for {{$labels.instance}}
-          description: "It is {{ $value }} connections for {{$labels.instance}}"
-invalid-field-2: invalid-value-2
-invalid-field-3: invalid-value-3

app/vmalert/config/testdata/rules/rules-multi-doc-duplicates-bad.rules

@@ -1,11 +0,0 @@
-groups:
-  - name: foo
-    rules:
-      - alert: VMRows
-        expr: vm_rows > 0
----
-groups:
-  - name: foo
-    rules:
-      - alert: VMRows
-        expr: vm_rows > 0

app/vmalert/config/testdata/rules/rules-multi-doc-good.rules

@@ -1,15 +0,0 @@
-
----
-groups:
-  - name: groupTest
-    rules:
-      - alert: VMRows
-        for: 1ms
-        expr: vm_rows > 0
-        labels:
-          label: bar
-          host: "{{ $labels.instance }}"
-        annotations:
-          summary: "{{ $value }}"
----
-groups:

app/vmalert/config/testdata/rules/rules-multi-doc2-good.rules

@@ -1,46 +0,0 @@
----
-groups:
-  - name: groupTest
-    rules:
-      - alert: VMRows
-        for: 1ms
-        expr: vm_rows > 0
-        labels:
-          label: bar
-          host: "{{ $labels.instance }}"
-        annotations:
-          summary: "{{ $value }}"
-  - name: groupTest-2
-    rules:
-      - alert: VMRows-2
-        for: 1ms
-        expr: vm_rows_2 > 0
-        labels:
-          label: bar2
-          host: "{{ $labels.instance }}"
-        annotations:
-          summary: "\n markdown result is : \n---\n # header\n body: \n text \n----\n"
----
-groups:
-  - name: groupTest-3
-    rules:
-      - alert: VMRows-3
-        for: 1ms
-        expr: vm_rows_3 > 0
-        labels:
-          label: bar_3
-          host: "{{ $labels.instance }}"
-        annotations:
-          summary: "{{ $value }}"
-  - name: groupTest-4
-    rules:
-      - alert: VMRows-4
-        for: 1ms
-        expr: vm_rows_4 > 0
-        labels:
-          label: bar4
-          host: "{{ $labels.instance }}"
-        annotations:
-          summary: "{{ $value }}"
----
-groups:

app/vmalert/config/testdata/rules/rules2-good.rules

@@ -22,7 +22,6 @@ groups:
               {{ . | first | value }}
             {{ end }}
           description: "It is {{ $value }} connections for {{$labels.instance}}"
-          link: http://localhost:3000/d/wNf0q_kZk?viewPanel=51&from={{($activeAt.Add (parseDurationTime "1h")).UnixMilli}}&to={{($activeAt.Add (parseDurationTime "-1h")).UnixMilli}}
       - alert: ExampleAlertAlwaysFiring
         update_entries_limit: -1
         expr: sum by(job)

app/vmalert/config/testdata/rules/rules3-good.rules

@@ -15,7 +15,6 @@ groups:
     interval: 2s
     concurrency: 2
     type: prometheus
-    eval_delay: 30s
     rules:
       - alert: Conns
         expr: sum(vm_tcplistener_conns) by (instance) > 1

app/vmalert/config/testdata/rules/rules_interval_bad.rules

@@ -1,13 +0,0 @@
-groups:
-  - name: groupTest
-    ## default interval is 1min, eval_offset shouldn't be greater than interval
-    eval_offset: 2m
-    rules:
-      - alert: VMRows
-        for: 2s
-        expr: sum(rate(vm_http_request_errors_total[2s])) > 0
-        labels:
-          label: bar
-          host: "{{ $labels.instance }}"
-        annotations:
-          summary: "{{ $value }}"

app/vmalert/config/testdata/rules/vlog-rules0-bad.rules

@@ -1,10 +0,0 @@
-groups:
-  - name: InvalidStatsLogsql
-    type: vlogs
-    interval: 5m
-    rules:
-      - record: MissingFilter
-        expr: 'stats count(*) as requests'
-      - record: MissingStatsPipe
-        expr: 'service: "nginx"'
-

app/vmalert/config/testdata/rules/vlog-rules0-good.rules

@@ -1,29 +0,0 @@
-groups:
-  - name: RequestCount
-    type: vlogs
-    interval: 5m
-    rules:
-      - record: nginxRequestCount
-        expr: 'env: "test" AND service: "nginx" | stats count(*) as requests'
-        annotations:
-          description: "Service nginx on env test accepted {{$labels.requests}} requests in the last 5 minutes"
-      - record: prodRequestCount
-        expr: 'env: "prod" | stats by (service) count(*) as requests'
-        annotations:
-          description: "Service {{$labels.service}} on env prod accepted {{$labels.requests}} requests in the last 5 minutes"
-  - name: ServiceLog
-    type: vlogs
-    interval: 5m
-    rules:
-      - alert: HasErrorLog
-        expr: 'env: "prod" AND status:~"error|warn" | stats by (service) count(*) as errorLog | filter errorLog:>0'
-        annotations:
-          description: "Service {{$labels.service}} generated {{$labels.errorLog}} error logs in the last 5 minutes"
-  - name: ServiceRequest
-    type: vlogs
-    interval: 10m
-    rules:
-      - alert: TooManyFailedRequest
-        expr: '* | extract "ip=<ip> " | extract "status_code=<code>;" | stats by (ip) count() if (code:!~200) as failed, count() as total| math failed / total as failed_percentage| filter failed_percentage :> 0.01 | fields ip,failed_percentage'
-        annotations:
-          description: "Connection from address {{$labels.ip}} has {{$value}} failed requests ratio in last 10 minutes"

app/vmalert/config/types.go

@@ -4,10 +4,8 @@ import (
 	"fmt"
 	"strings"
 
-	"github.com/VictoriaMetrics/VictoriaLogs/lib/logstorage"
-	"github.com/VictoriaMetrics/metricsql"
-
 	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmselect/graphiteql"
+	"github.com/VictoriaMetrics/metricsql"
 )
 
 // Type represents data source type
@@ -29,13 +27,6 @@ func NewGraphiteType() Type {
 	}
 }
 
-// NewVLogsType returns victorialogs datasource type
-func NewVLogsType() Type {
-	return Type{
-		Name: "vlogs",
-	}
-}
-
 // NewRawType returns datasource type from raw string
 // without validation.
 func NewRawType(d string) Type {
@@ -71,45 +62,32 @@ func (t *Type) ValidateExpr(expr string) error {
 		if _, err := metricsql.Parse(expr); err != nil {
 			return fmt.Errorf("bad prometheus expr: %q, err: %w", expr, err)
 		}
-	case "vlogs":
-		q, err := logstorage.ParseStatsQuery(expr, 0)
-		if err != nil {
-			return fmt.Errorf("bad LogsQL expr: %q, err: %w", expr, err)
-		}
-		fields, _ := q.GetStatsByFields()
-		for i := range fields {
-			// VictoriaLogs inserts `_time` field as a label in result when query with `stats by (_time:step)`,
-			// making the result meaningless and may lead to cardinality issues.
-			if fields[i] == "_time" {
-				return fmt.Errorf("bad LogsQL expr: %q, err: cannot contain time buckets stats pipe `stats by (_time:step)`", expr)
-			}
-		}
 	default:
 		return fmt.Errorf("unknown datasource type=%q", t.Name)
 	}
 	return nil
 }
 
-// SupportedType is true if given datasource type is supported
-func SupportedType(dsType string) bool {
-	return dsType == "graphite" || dsType == "prometheus" || dsType == "vlogs"
-}
-
 // UnmarshalYAML implements the yaml.Unmarshaler interface.
-func (t *Type) UnmarshalYAML(unmarshal func(any) error) error {
+func (t *Type) UnmarshalYAML(unmarshal func(interface{}) error) error {
 	var s string
 	if err := unmarshal(&s); err != nil {
 		return err
 	}
-	if !SupportedType(s) {
-		return fmt.Errorf("unknown datasource type=%q, want prometheus, graphite or vlogs", s)
+	if s == "" {
+		s = "prometheus"
+	}
+	switch s {
+	case "graphite", "prometheus":
+	default:
+		return fmt.Errorf("unknown datasource type=%q, want %q or %q", s, "prometheus", "graphite")
 	}
 	t.Name = s
 	return nil
 }
 
 // MarshalYAML implements the yaml.Unmarshaler interface.
-func (t Type) MarshalYAML() (any, error) {
+func (t Type) MarshalYAML() (interface{}, error) {
 	return t.Name, nil
 }
 
@@ -120,7 +98,7 @@ type Header struct {
 }
 
 // UnmarshalYAML implements the yaml.Unmarshaler interface.
-func (h *Header) UnmarshalYAML(unmarshal func(any) error) error {
+func (h *Header) UnmarshalYAML(unmarshal func(interface{}) error) error {
 	var s string
 	if err := unmarshal(&s); err != nil {
 		return err

app/vmalert/datasource/client.go

@@ -1,333 +0,0 @@
-package datasource
-
-import (
-	"context"
-	"errors"
-	"fmt"
-	"io"
-	"net/http"
-	"net/url"
-	"strings"
-	"time"
-
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/netutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promauth"
-)
-
-type datasourceType string
-
-const (
-	datasourcePrometheus datasourceType = "prometheus"
-	datasourceGraphite   datasourceType = "graphite"
-	datasourceVLogs      datasourceType = "vlogs"
-)
-
-func toDatasourceType(s string) datasourceType {
-	switch s {
-	case string(datasourcePrometheus):
-		return datasourcePrometheus
-	case string(datasourceGraphite):
-		return datasourceGraphite
-	case string(datasourceVLogs):
-		return datasourceVLogs
-	default:
-		logger.Panicf("BUG: unknown datasource type %q", s)
-	}
-	return ""
-}
-
-// Client is a datasource entity for reading data,
-// supported clients are enumerated in datasourceType.
-// WARN: when adding a new field, remember to check if Clone() method needs to be updated.
-type Client struct {
-	c                *http.Client
-	authCfg          *promauth.Config
-	datasourceURL    string
-	appendTypePrefix bool
-	queryStep        time.Duration
-	dataSourceType   datasourceType
-	// ApplyIntervalAsTimeFilter is only valid for vlogs datasource.
-	// Set to true if there is no [timeFilter](https://docs.victoriametrics.com/victorialogs/logsql/#time-filter) in the rule expression,
-	// and we will add evaluation interval as an additional timeFilter when querying.
-	applyIntervalAsTimeFilter bool
-
-	// evaluationInterval will help setting request's `step` param,
-	// or adding time filter for LogsQL expression.
-	evaluationInterval time.Duration
-	// extraParams contains params to be attached to each HTTP request
-	extraParams url.Values
-	// extraHeaders are headers to be attached to each HTTP request
-	extraHeaders []keyValue
-
-	// whether to print additional log messages
-	// for each sent request
-	debug bool
-}
-
-type keyValue struct {
-	key   string
-	value string
-}
-
-// Clone clones shared http client and other configuration to the new client.
-func (c *Client) Clone() *Client {
-	ns := &Client{
-		c:                c.c,
-		authCfg:          c.authCfg,
-		datasourceURL:    c.datasourceURL,
-		appendTypePrefix: c.appendTypePrefix,
-		queryStep:        c.queryStep,
-
-		dataSourceType:     c.dataSourceType,
-		evaluationInterval: c.evaluationInterval,
-
-		// init map so it can be populated below
-		extraParams: url.Values{},
-
-		debug: c.debug,
-	}
-	if len(c.extraHeaders) > 0 {
-		ns.extraHeaders = make([]keyValue, len(c.extraHeaders))
-		copy(ns.extraHeaders, c.extraHeaders)
-	}
-	for k, v := range c.extraParams {
-		ns.extraParams[k] = v
-	}
-
-	return ns
-}
-
-// ApplyParams - changes given querier params.
-func (c *Client) ApplyParams(params QuerierParams) *Client {
-	if params.DataSourceType != "" {
-		c.dataSourceType = toDatasourceType(params.DataSourceType)
-	}
-	c.evaluationInterval = params.EvaluationInterval
-	c.applyIntervalAsTimeFilter = params.ApplyIntervalAsTimeFilter
-	if params.QueryParams != nil {
-		if c.extraParams == nil {
-			c.extraParams = url.Values{}
-		}
-		for k, vl := range params.QueryParams {
-			// custom query params are prior to default ones
-			if c.extraParams.Has(k) {
-				c.extraParams.Del(k)
-			}
-			for _, v := range vl {
-				// don't use .Set() instead of Del/Add since it is allowed
-				// for GET params to be duplicated
-				// see https://github.com/VictoriaMetrics/VictoriaMetrics/issues/4908
-				c.extraParams.Add(k, v)
-			}
-		}
-	}
-	if params.Headers != nil {
-		for key, value := range params.Headers {
-			kv := keyValue{key: key, value: value}
-			c.extraHeaders = append(c.extraHeaders, kv)
-		}
-	}
-	c.debug = params.Debug
-	return c
-}
-
-// BuildWithParams - implements interface.
-func (c *Client) BuildWithParams(params QuerierParams) Querier {
-	return c.Clone().ApplyParams(params)
-}
-
-// NewPrometheusClient returns a new prometheus datasource client.
-func NewPrometheusClient(baseURL string, authCfg *promauth.Config, appendTypePrefix bool, c *http.Client) *Client {
-	return &Client{
-		c:                c,
-		authCfg:          authCfg,
-		datasourceURL:    strings.TrimSuffix(baseURL, "/"),
-		appendTypePrefix: appendTypePrefix,
-		queryStep:        *queryStep,
-		dataSourceType:   datasourcePrometheus,
-		extraParams:      url.Values{},
-	}
-}
-
-// Query executes the given query and returns parsed response
-func (c *Client) Query(ctx context.Context, query string, ts time.Time) (Result, *http.Request, error) {
-	req, err := c.newQueryRequest(ctx, query, ts)
-	if err != nil {
-		return Result{}, nil, err
-	}
-	resp, err := c.do(req)
-	if err != nil {
-		if !errors.Is(err, io.EOF) && !errors.Is(err, io.ErrUnexpectedEOF) && !netutil.IsTrivialNetworkError(err) {
-			// Return unexpected error to the caller.
-			return Result{}, nil, err
-		}
-		// Something in the middle between client and datasource might be closing
-		// the connection. So we do a one more attempt in hope request will succeed.
-		req, err = c.newQueryRequest(ctx, query, ts)
-		if err != nil {
-			return Result{}, nil, fmt.Errorf("second attempt: %w", err)
-		}
-		resp, err = c.do(req)
-		if err != nil {
-			return Result{}, nil, fmt.Errorf("second attempt: %w", err)
-		}
-	}
-
-	// Process the received response.
-	var parseFn func(req *http.Request, resp *http.Response) (Result, error)
-	switch c.dataSourceType {
-	case datasourcePrometheus:
-		parseFn = parsePrometheusResponse
-	case datasourceGraphite:
-		parseFn = parseGraphiteResponse
-	case datasourceVLogs:
-		parseFn = parseVLogsResponse
-	default:
-		logger.Panicf("BUG: unsupported datasource type %q to parse query response", c.dataSourceType)
-	}
-	result, err := parseFn(req, resp)
-	_ = resp.Body.Close()
-	return result, req, err
-}
-
-// QueryRange executes the given query on the given time range.
-// For Prometheus type see https://prometheus.io/docs/prometheus/latest/querying/api/#range-queries
-// Graphite type isn't supported.
-func (c *Client) QueryRange(ctx context.Context, query string, start, end time.Time) (res Result, err error) {
-	if c.dataSourceType == datasourceGraphite {
-		return res, fmt.Errorf("%q is not supported for QueryRange", c.dataSourceType)
-	}
-	// TODO: disable range query LogsQL with time filter now
-	if c.dataSourceType == datasourceVLogs && !c.applyIntervalAsTimeFilter {
-		return res, fmt.Errorf("range query is not supported for LogsQL expression %q because it contains time filter. Remove time filter from the expression and try again", query)
-	}
-	if start.IsZero() {
-		return res, fmt.Errorf("start param is missing")
-	}
-	if end.IsZero() {
-		return res, fmt.Errorf("end param is missing")
-	}
-	req, err := c.newQueryRangeRequest(ctx, query, start, end)
-	if err != nil {
-		return res, err
-	}
-	resp, err := c.do(req)
-	if err != nil {
-		if !errors.Is(err, io.EOF) && !errors.Is(err, io.ErrUnexpectedEOF) && !netutil.IsTrivialNetworkError(err) {
-			// Return unexpected error to the caller.
-			return res, err
-		}
-		// Something in the middle between client and datasource might be closing
-		// the connection. So we do a one more attempt in hope request will succeed.
-		req, err = c.newQueryRangeRequest(ctx, query, start, end)
-		if err != nil {
-			return res, fmt.Errorf("second attempt: %w", err)
-		}
-		resp, err = c.do(req)
-		if err != nil {
-			return res, fmt.Errorf("second attempt: %w", err)
-		}
-	}
-
-	// Process the received response.
-	var parseFn func(req *http.Request, resp *http.Response) (Result, error)
-	switch c.dataSourceType {
-	case datasourcePrometheus:
-		parseFn = parsePrometheusResponse
-	case datasourceVLogs:
-		parseFn = parseVLogsResponse
-	default:
-		logger.Panicf("BUG: unsupported datasource type %q to parse query range response", c.dataSourceType)
-	}
-	res, err = parseFn(req, resp)
-	_ = resp.Body.Close()
-	return res, err
-}
-
-func (c *Client) do(req *http.Request) (*http.Response, error) {
-	ru := req.URL.Redacted()
-	if *showDatasourceURL {
-		ru = req.URL.String()
-	}
-	if c.debug {
-		logger.Infof("DEBUG datasource request: executing %s request with params %q", req.Method, ru)
-	}
-	resp, err := c.c.Do(req)
-	if err != nil {
-		return nil, fmt.Errorf("error getting response from %s: %w", ru, err)
-	}
-	if resp.StatusCode != http.StatusOK {
-		body, _ := io.ReadAll(resp.Body)
-		_ = resp.Body.Close()
-		return nil, fmt.Errorf("unexpected response code %d for %s. Response body %s", resp.StatusCode, ru, body)
-	}
-	return resp, nil
-}
-
-func (c *Client) newQueryRangeRequest(ctx context.Context, query string, start, end time.Time) (*http.Request, error) {
-	req, err := c.newRequest(ctx)
-	if err != nil {
-		return nil, fmt.Errorf("cannot create query_range request to datasource %q: %w", c.datasourceURL, err)
-	}
-	switch c.dataSourceType {
-	case datasourcePrometheus:
-		c.setPrometheusRangeReqParams(req, query, start, end)
-	case datasourceVLogs:
-		c.setVLogsRangeReqParams(req, query, start, end)
-	default:
-		logger.Panicf("BUG: unsupported datasource type %q to create range query request", c.dataSourceType)
-	}
-	return req, nil
-}
-
-func (c *Client) newQueryRequest(ctx context.Context, query string, ts time.Time) (*http.Request, error) {
-	req, err := c.newRequest(ctx)
-	if err != nil {
-		return nil, fmt.Errorf("cannot create query request to datasource %q: %w", c.datasourceURL, err)
-	}
-	switch c.dataSourceType {
-	case datasourcePrometheus:
-		c.setPrometheusInstantReqParams(req, query, ts)
-	case datasourceGraphite:
-		c.setGraphiteReqParams(req, query)
-	case datasourceVLogs:
-		c.setVLogsInstantReqParams(req, query, ts)
-	default:
-		logger.Panicf("BUG: unsupported datasource type %q to create query request", c.dataSourceType)
-	}
-	return req, nil
-}
-
-func (c *Client) newRequest(ctx context.Context) (*http.Request, error) {
-	req, err := http.NewRequestWithContext(ctx, http.MethodPost, c.datasourceURL, nil)
-	if err != nil {
-		logger.Panicf("BUG: unexpected error from http.NewRequest(%q): %s", c.datasourceURL, err)
-	}
-	req.Header.Set("Content-Type", "application/json")
-	if c.authCfg != nil {
-		err = c.authCfg.SetHeaders(req, true)
-		if err != nil {
-			return nil, err
-		}
-	}
-	for _, h := range c.extraHeaders {
-		req.Header.Set(h.key, h.value)
-	}
-	return req, nil
-}
-
-// setReqParams adds query and other extra params for the request.
-func (c *Client) setReqParams(r *http.Request, query string) {
-	q := r.URL.Query()
-	for k, vs := range c.extraParams {
-		if q.Has(k) { // extraParams are prior to params in URL
-			q.Del(k)
-		}
-		for _, v := range vs {
-			q.Add(k, v)
-		}
-	}
-	q.Set("query", query)
-	r.URL.RawQuery = q.Encode()
-}

app/vmalert/datasource/client_graphite.go

@@ -1,71 +0,0 @@
-package datasource
-
-import (
-	"encoding/json"
-	"fmt"
-	"net/http"
-)
-
-type graphiteResponse []graphiteResponseTarget
-
-type graphiteResponseTarget struct {
-	Target     string            `json:"target"`
-	Tags       map[string]string `json:"tags"`
-	DataPoints [][2]float64      `json:"datapoints"`
-}
-
-func (r graphiteResponse) metrics() []Metric {
-	var ms []Metric
-	for _, res := range r {
-		if len(res.DataPoints) < 1 {
-			continue
-		}
-		var m Metric
-		// add only last value to the result.
-		last := res.DataPoints[len(res.DataPoints)-1]
-		m.Values = append(m.Values, last[0])
-		m.Timestamps = append(m.Timestamps, int64(last[1]))
-		for k, v := range res.Tags {
-			m.AddLabel(k, v)
-		}
-		ms = append(ms, m)
-	}
-	return ms
-}
-
-func parseGraphiteResponse(req *http.Request, resp *http.Response) (Result, error) {
-	r := &graphiteResponse{}
-	if err := json.NewDecoder(resp.Body).Decode(r); err != nil {
-		return Result{}, fmt.Errorf("error parsing graphite metrics for %s: %w", req.URL.Redacted(), err)
-	}
-	return Result{Data: r.metrics()}, nil
-}
-
-const (
-	graphitePath   = "/render"
-	graphitePrefix = "/graphite"
-)
-
-func (c *Client) setGraphiteReqParams(r *http.Request, query string) {
-	if c.appendTypePrefix {
-		r.URL.Path += graphitePrefix
-	}
-	r.URL.Path += graphitePath
-	q := r.URL.Query()
-	from := "-5min"
-	q.Set("from", from)
-	q.Set("format", "json")
-	q.Set("target", query)
-	q.Set("until", "now")
-
-	for k, vs := range c.extraParams {
-		if q.Has(k) { // extraParams are prior to params in URL
-			q.Del(k)
-		}
-		for _, v := range vs {
-			q.Add(k, v)
-		}
-	}
-
-	r.URL.RawQuery = q.Encode()
-}

app/vmalert/datasource/client_prom.go

@@ -1,265 +0,0 @@
-package datasource
-
-import (
-	"encoding/json"
-	"flag"
-	"fmt"
-	"net/http"
-	"strconv"
-	"time"
-
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
-
-	"github.com/valyala/fastjson"
-)
-
-var (
-	disablePathAppend = flag.Bool("remoteRead.disablePathAppend", false, "Whether to disable automatic appending of '/api/v1/query' or '/select/logsql/stats_query' path "+
-		"to the configured -datasource.url and -remoteRead.url")
-	disableStepParam = flag.Bool("datasource.disableStepParam", false, "Whether to disable adding 'step' param in instant queries to the configured -datasource.url and -remoteRead.url. "+
-		"Only valid for prometheus datasource. "+
-		"This might be useful when using vmalert with datasources that do not support 'step' param for instant queries, like Google Managed Prometheus. "+
-		"It is not recommended to enable this flag if you use vmalert with VictoriaMetrics.")
-)
-
-type promResponse struct {
-	Status    string `json:"status"`
-	ErrorType string `json:"errorType"`
-	Error     string `json:"error"`
-	Data      struct {
-		ResultType string          `json:"resultType"`
-		Result     json.RawMessage `json:"result"`
-	} `json:"data"`
-	// Stats supported by VictoriaMetrics since v1.90
-	Stats struct {
-		SeriesFetched *string `json:"seriesFetched,omitempty"`
-	} `json:"stats,omitempty"`
-	// IsPartial supported by VictoriaMetrics
-	IsPartial *bool `json:"isPartial,omitempty"`
-}
-
-// see https://prometheus.io/docs/prometheus/latest/querying/api/#instant-queries
-type promInstant struct {
-	// ms is populated after Unmarshal call
-	ms []Metric
-}
-
-// metrics returned parsed Metric slice
-// Must be called only after Unmarshal
-func (pi *promInstant) metrics() ([]Metric, error) {
-	return pi.ms, nil
-}
-
-var jsonParserPool fastjson.ParserPool
-
-// Unmarshal unmarshals the given byte slice into promInstant
-// It is using fastjson to reduce number of allocations compared to
-// standard json.Unmarshal function.
-// Response example:
-//
-//	[{"metric":{"__name__":"up","job":"prometheus"},value": [ 1435781451.781,"1"]},
-//	{"metric":{"__name__":"up","job":"node"},value": [ 1435781451.781,"0"]}]
-func (pi *promInstant) Unmarshal(b []byte) error {
-	var metrics []json.RawMessage
-	// metrics slice could be large, so parsing it with fastjson could consume a lot of memory.
-	// We parse the slice with standard lib to keep mem usage low.
-	// And each metric object will be parsed with fastjson to reduce allocations.
-	if err := json.Unmarshal(b, &metrics); err != nil {
-		return fmt.Errorf("cannot unmarshal metrics: %w", err)
-	}
-
-	p := jsonParserPool.Get()
-	defer jsonParserPool.Put(p)
-
-	pi.ms = make([]Metric, len(metrics))
-	for i, data := range metrics {
-		row, err := p.ParseBytes(data)
-		if err != nil {
-			return fmt.Errorf("cannot parse metric object: %w", err)
-		}
-		metric := row.Get("metric")
-		if metric == nil {
-			return fmt.Errorf("can't find `metric` object in %q", row)
-		}
-		labels := metric.GetObject()
-
-		r := &pi.ms[i]
-		r.Labels = make([]prompb.Label, 0, labels.Len())
-		labels.Visit(func(key []byte, v *fastjson.Value) {
-			lv, errLocal := v.StringBytes()
-			if errLocal != nil {
-				err = fmt.Errorf("error when parsing label value %q: %s", v, errLocal)
-				return
-			}
-			r.Labels = append(r.Labels, prompb.Label{
-				Name:  string(key),
-				Value: string(lv),
-			})
-		})
-		if err != nil {
-			return fmt.Errorf("error when parsing `metric` object in %q: %w", row, err)
-		}
-
-		value := row.Get("value")
-		if value == nil {
-			return fmt.Errorf("can't find `value` object in %q", row)
-		}
-		sample := value.GetArray()
-		if len(sample) != 2 {
-			return fmt.Errorf("object `value` in %q should contain 2 values, but contains %d instead", row, len(sample))
-		}
-		r.Timestamps = []int64{sample[0].GetInt64()}
-		val, err := sample[1].StringBytes()
-		if err != nil {
-			return fmt.Errorf("error when parsing `value` object %q: %s", sample[1], err)
-		}
-		f, err := strconv.ParseFloat(bytesutil.ToUnsafeString(val), 64)
-		if err != nil {
-			return fmt.Errorf("error when parsing float64 from %s in %q: %w", sample[1], row, err)
-		}
-		r.Values = []float64{f}
-	}
-	return nil
-}
-
-type promRange struct {
-	Result []struct {
-		Labels map[string]string `json:"metric"`
-		TVs    [][2]any          `json:"values"`
-	} `json:"result"`
-}
-
-func (r promRange) metrics() ([]Metric, error) {
-	var result []Metric
-	for i, res := range r.Result {
-		var m Metric
-		for _, tv := range res.TVs {
-			f, err := strconv.ParseFloat(tv[1].(string), 64)
-			if err != nil {
-				return nil, fmt.Errorf("metric %v, unable to parse float64 from %s: %w", res, tv[1], err)
-			}
-			m.Values = append(m.Values, f)
-			m.Timestamps = append(m.Timestamps, int64(tv[0].(float64)))
-		}
-		if len(m.Values) < 1 || len(m.Timestamps) < 1 {
-			return nil, fmt.Errorf("metric %v contains no values", res)
-		}
-		m.Labels = nil
-		for k, v := range r.Result[i].Labels {
-			m.AddLabel(k, v)
-		}
-		result = append(result, m)
-	}
-	return result, nil
-}
-
-type promScalar [2]any
-
-func (r promScalar) metrics() ([]Metric, error) {
-	var m Metric
-	f, err := strconv.ParseFloat(r[1].(string), 64)
-	if err != nil {
-		return nil, fmt.Errorf("metric %v, unable to parse float64 from %s: %w", r, r[1], err)
-	}
-	m.Values = append(m.Values, f)
-	m.Timestamps = append(m.Timestamps, int64(r[0].(float64)))
-	return []Metric{m}, nil
-}
-
-const (
-	statusSuccess, statusError  = "success", "error"
-	rtVector, rtMatrix, rScalar = "vector", "matrix", "scalar"
-)
-
-func parsePrometheusResponse(req *http.Request, resp *http.Response) (res Result, err error) {
-	r := &promResponse{}
-	if err = json.NewDecoder(resp.Body).Decode(r); err != nil {
-		return res, fmt.Errorf("error parsing response from %s: %w", req.URL.Redacted(), err)
-	}
-	if r.Status == statusError {
-		return res, fmt.Errorf("response error, query: %s, errorType: %s, error: %s", req.URL.Redacted(), r.ErrorType, r.Error)
-	}
-	if r.Status != statusSuccess {
-		return res, fmt.Errorf("unknown status: %s, Expected success or error", r.Status)
-	}
-	var parseFn func() ([]Metric, error)
-	switch r.Data.ResultType {
-	case rtVector:
-		var pi promInstant
-		if err := pi.Unmarshal(r.Data.Result); err != nil {
-			return res, fmt.Errorf("unmarshal err %w; \n %#v", err, string(r.Data.Result))
-		}
-		parseFn = pi.metrics
-	case rtMatrix:
-		var pr promRange
-		if err := json.Unmarshal(r.Data.Result, &pr.Result); err != nil {
-			return res, err
-		}
-		parseFn = pr.metrics
-	case rScalar:
-		var ps promScalar
-		if err := json.Unmarshal(r.Data.Result, &ps); err != nil {
-			return res, err
-		}
-		parseFn = ps.metrics
-	default:
-		return res, fmt.Errorf("unknown result type %q", r.Data.ResultType)
-	}
-
-	ms, err := parseFn()
-	if err != nil {
-		return res, err
-	}
-	res = Result{Data: ms, IsPartial: r.IsPartial}
-	if r.Stats.SeriesFetched != nil {
-		intV, err := strconv.Atoi(*r.Stats.SeriesFetched)
-		if err != nil {
-			return res, fmt.Errorf("failed to convert stats.seriesFetched to int: %w", err)
-		}
-		res.SeriesFetched = &intV
-	}
-	return res, nil
-}
-
-func (c *Client) setPrometheusInstantReqParams(r *http.Request, query string, timestamp time.Time) {
-	if c.appendTypePrefix {
-		r.URL.Path += "/prometheus"
-	}
-	if !*disablePathAppend {
-		r.URL.Path += "/api/v1/query"
-	}
-	q := r.URL.Query()
-	q.Set("time", timestamp.Format(time.RFC3339))
-	if !*disableStepParam && c.evaluationInterval > 0 { // set step as evaluationInterval by default
-		// always convert to seconds to keep compatibility with older
-		// Prometheus versions. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1943
-		q.Set("step", fmt.Sprintf("%ds", int(c.evaluationInterval.Seconds())))
-	}
-	if !*disableStepParam && c.queryStep > 0 { // override step with user-specified value
-		// always convert to seconds to keep compatibility with older
-		// Prometheus versions. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1943
-		q.Set("step", fmt.Sprintf("%ds", int(c.queryStep.Seconds())))
-	}
-	r.URL.RawQuery = q.Encode()
-	c.setReqParams(r, query)
-}
-
-func (c *Client) setPrometheusRangeReqParams(r *http.Request, query string, start, end time.Time) {
-	if c.appendTypePrefix {
-		r.URL.Path += "/prometheus"
-	}
-	if !*disablePathAppend {
-		r.URL.Path += "/api/v1/query_range"
-	}
-	q := r.URL.Query()
-	q.Add("start", start.Format(time.RFC3339))
-	q.Add("end", end.Format(time.RFC3339))
-	if c.evaluationInterval > 0 { // set step as evaluationInterval by default
-		// always convert to seconds to keep compatibility with older
-		// Prometheus versions. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1943
-		q.Set("step", fmt.Sprintf("%ds", int(c.evaluationInterval.Seconds())))
-	}
-	r.URL.RawQuery = q.Encode()
-	c.setReqParams(r, query)
-}

app/vmalert/datasource/client_test.go

@@ -1,844 +0,0 @@
-package datasource
-
-import (
-	"context"
-	"fmt"
-	"net/http"
-	"net/http/httptest"
-	"net/url"
-	"reflect"
-	"sort"
-	"strings"
-	"testing"
-	"time"
-
-	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/vmalertutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promauth"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
-)
-
-var (
-	ctx           = context.Background()
-	basicAuthName = "foo"
-	basicAuthPass = "bar"
-	baCfg         = &promauth.BasicAuthConfig{
-		Username: basicAuthName,
-		Password: promauth.NewSecret(basicAuthPass),
-	}
-	vmQuery         = "vm_rows"
-	queryRender     = "constantLine(10)"
-	vlogsQuery      = "_time: 5m | stats by (foo) count() total"
-	vlogsRangeQuery = "* | stats by (foo) count() total"
-)
-
-func TestVMInstantQuery(t *testing.T) {
-	mux := http.NewServeMux()
-	mux.HandleFunc("/", func(_ http.ResponseWriter, _ *http.Request) {
-		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.Fatalf("expected POST method got %s", r.Method)
-		}
-		if name, pass, _ := r.BasicAuth(); name != basicAuthName || pass != basicAuthPass {
-			t.Fatalf("expected %s:%s as basic auth got %s:%s", basicAuthName, basicAuthPass, name, pass)
-		}
-		if r.URL.Query().Get("query") != vmQuery {
-			t.Fatalf("expected %s in query param, got %s", vmQuery, r.URL.Query().Get("query"))
-		}
-		timeParam := r.URL.Query().Get("time")
-		if timeParam == "" {
-			t.Fatalf("expected 'time' in query param, got nil instead")
-		}
-		if _, err := time.Parse(time.RFC3339, timeParam); err != nil {
-			t.Fatalf("failed to parse 'time' query param %q: %s", timeParam, err)
-		}
-		switch c {
-		case 0:
-			w.WriteHeader(500)
-		case 1:
-			w.Write([]byte("[]"))
-		case 2:
-			w.Write([]byte(`{"status":"error", "errorType":"type:", "error":"some error msg"}`))
-		case 3:
-			w.Write([]byte(`{"status":"unknown"}`))
-		case 4:
-			w.Write([]byte(`{"status":"success","data":{"resultType":"matrix"}}`))
-		case 5:
-			w.Write([]byte(`{"status":"success","data":{"resultType":"vector","result":[{"metric":{"__name__":"vm_rows","foo":"bar"},"value":[1583786142,"13763"]},{"metric":{"__name__":"vm_requests","foo":"baz"},"value":[1583786140,"2000"]}]}}`))
-		case 6:
-			w.Write([]byte(`{"status":"success","data":{"resultType":"scalar","result":[1583786142, "1"]}}`))
-		case 7:
-			w.Write([]byte(`{"status":"success","data":{"resultType":"scalar","result":[1583786142, "1"]},"stats":{"seriesFetched": "42"}}`))
-		case 8:
-			w.Write([]byte(`{"status":"success", "isPartial":true, "data":{"resultType":"scalar","result":[1583786142, "1"]}}`))
-		}
-	})
-	mux.HandleFunc("/render", func(w http.ResponseWriter, _ *http.Request) {
-		c++
-		switch c {
-		case 9:
-			w.Write([]byte(`[{"target":"constantLine(10)","tags":{"name":"constantLine(10)"},"datapoints":[[10,1611758343],[10,1611758373],[10,1611758403]]}]`))
-		}
-	})
-	mux.HandleFunc("/select/logsql/stats_query", func(w http.ResponseWriter, r *http.Request) {
-		c++
-		if r.Method != http.MethodPost {
-			t.Fatalf("expected POST method got %s", r.Method)
-		}
-		if name, pass, _ := r.BasicAuth(); name != basicAuthName || pass != basicAuthPass {
-			t.Fatalf("expected %s:%s as basic auth got %s:%s", basicAuthName, basicAuthPass, name, pass)
-		}
-		if r.URL.Query().Get("query") != vlogsQuery {
-			t.Fatalf("expected %s in query param, got %s", vlogsQuery, r.URL.Query().Get("query"))
-		}
-		timeParam := r.URL.Query().Get("time")
-		if timeParam == "" {
-			t.Fatalf("expected 'time' in query param, got nil instead")
-		}
-		if _, err := time.Parse(time.RFC3339, timeParam); err != nil {
-			t.Fatalf("failed to parse 'time' query param %q: %s", timeParam, err)
-		}
-		switch c {
-		case 10:
-			w.Write([]byte("[]"))
-		case 11:
-			w.Write([]byte(`{"status":"success","data":{"resultType":"vector","result":[{"metric":{"__name__":"total","foo":"bar"},"value":[1583786142,"13763"]},{"metric":{"__name__":"total","foo":"baz"},"value":[1583786140,"2000"]}]}}`))
-		}
-	})
-
-	srv := httptest.NewServer(mux)
-	defer srv.Close()
-
-	authCfg, err := baCfg.NewConfig(".")
-	if err != nil {
-		t.Fatalf("unexpected: %s", err)
-	}
-	s := NewPrometheusClient(srv.URL, authCfg, false, srv.Client())
-
-	p := datasourcePrometheus
-	pq := s.BuildWithParams(QuerierParams{DataSourceType: string(p), EvaluationInterval: 15 * time.Second})
-	ts := time.Now()
-
-	expErr := func(query, err string) {
-		_, _, gotErr := pq.Query(ctx, query, ts)
-		if gotErr == nil {
-			t.Fatalf("expected %q got nil", err)
-		}
-		if !strings.Contains(gotErr.Error(), err) {
-			t.Fatalf("expected err %q; got %q", err, gotErr)
-		}
-	}
-
-	expErr(vmQuery, "500")                          // 0
-	expErr(vmQuery, "error parsing response")       // 1
-	expErr(vmQuery, "response error")               // 2
-	expErr(vmQuery, "unknown status")               // 3
-	expErr(vmQuery, "unexpected end of JSON input") // 4
-
-	res, _, err := pq.Query(ctx, vmQuery, ts) // 5 - vector
-	if err != nil {
-		t.Fatalf("unexpected %s", err)
-	}
-	if len(res.Data) != 2 {
-		t.Fatalf("expected 2 metrics got %d in %+v", len(res.Data), res.Data)
-	}
-	expected := []Metric{
-		{
-			Labels:     []prompb.Label{{Value: "vm_rows", Name: "__name__"}, {Value: "bar", Name: "foo"}},
-			Timestamps: []int64{1583786142},
-			Values:     []float64{13763},
-		},
-		{
-			Labels:     []prompb.Label{{Value: "vm_requests", Name: "__name__"}, {Value: "baz", Name: "foo"}},
-			Timestamps: []int64{1583786140},
-			Values:     []float64{2000},
-		},
-	}
-	metricsEqual(t, res.Data, expected)
-
-	res, req, err := pq.Query(ctx, vmQuery, ts) // 6 - scalar
-	if err != nil {
-		t.Fatalf("unexpected %s", err)
-	}
-	if req == nil {
-		t.Fatalf("expected request to be non-nil")
-	}
-	if len(res.Data) != 1 {
-		t.Fatalf("expected 1 metrics got %d in %+v", len(res.Data), res.Data)
-	}
-	expected = []Metric{
-		{
-			Timestamps: []int64{1583786142},
-			Values:     []float64{1},
-		},
-	}
-	if !reflect.DeepEqual(res.Data, expected) {
-		t.Fatalf("unexpected metric %+v want %+v", res.Data, expected)
-	}
-
-	if res.SeriesFetched != nil {
-		t.Fatalf("expected `seriesFetched` field to be nil when it is missing in datasource response; got %v instead",
-			res.SeriesFetched)
-	}
-
-	res, _, err = pq.Query(ctx, vmQuery, ts) // 7 - scalar with stats
-	if err != nil {
-		t.Fatalf("unexpected %s", err)
-	}
-	if len(res.Data) != 1 {
-		t.Fatalf("expected 1 metrics got %d in %+v", len(res.Data), res)
-	}
-	expected = []Metric{
-		{
-			Timestamps: []int64{1583786142},
-			Values:     []float64{1},
-		},
-	}
-	if !reflect.DeepEqual(res.Data, expected) {
-		t.Fatalf("unexpected metric %+v want %+v", res.Data, expected)
-	}
-	if *res.SeriesFetched != 42 {
-		t.Fatalf("expected `seriesFetched` field to be 42; got %d instead",
-			*res.SeriesFetched)
-	}
-
-	res, _, err = pq.Query(ctx, vmQuery, ts) // 8
-	if err != nil {
-		t.Fatalf("unexpected %s", err)
-	}
-	if res.IsPartial != nil && !*res.IsPartial {
-		t.Fatalf("unexpected metric isPartial want %+v", true)
-	}
-
-	// test graphite
-	gq := s.BuildWithParams(QuerierParams{DataSourceType: string(datasourceGraphite)})
-
-	res, _, err = gq.Query(ctx, queryRender, ts) // 9 - graphite
-	if err != nil {
-		t.Fatalf("unexpected %s", err)
-	}
-	if len(res.Data) != 1 {
-		t.Fatalf("expected 1 metric  got %d in %+v", len(res.Data), res.Data)
-	}
-	exp := []Metric{
-		{
-			Labels:     []prompb.Label{{Value: "constantLine(10)", Name: "name"}},
-			Timestamps: []int64{1611758403},
-			Values:     []float64{10},
-		},
-	}
-	metricsEqual(t, res.Data, exp)
-
-	// test victorialogs
-	vlogs := datasourceVLogs
-	pq = s.BuildWithParams(QuerierParams{DataSourceType: string(vlogs), EvaluationInterval: 15 * time.Second})
-
-	expErr(vlogsQuery, "error parsing response") // 10
-
-	res, _, err = pq.Query(ctx, vlogsQuery, ts) // 11
-	if err != nil {
-		t.Fatalf("unexpected %s", err)
-	}
-	if len(res.Data) != 2 {
-		t.Fatalf("expected 2 metrics got %d in %+v", len(res.Data), res.Data)
-	}
-	expected = []Metric{
-		{
-			Labels:     []prompb.Label{{Value: "total", Name: "stats_result"}, {Value: "bar", Name: "foo"}},
-			Timestamps: []int64{1583786142},
-			Values:     []float64{13763},
-		},
-		{
-			Labels:     []prompb.Label{{Value: "total", Name: "stats_result"}, {Value: "baz", Name: "foo"}},
-			Timestamps: []int64{1583786140},
-			Values:     []float64{2000},
-		},
-	}
-	metricsEqual(t, res.Data, expected)
-}
-
-func TestVMInstantQueryWithRetry(t *testing.T) {
-	mux := http.NewServeMux()
-	mux.HandleFunc("/", func(_ http.ResponseWriter, _ *http.Request) {
-		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") != vmQuery {
-			t.Fatalf("expected %s in query param, got %s", vmQuery, r.URL.Query().Get("query"))
-		}
-		switch c {
-		case 0:
-			w.Write([]byte(`{"status":"success","data":{"resultType":"scalar","result":[1583786142, "1"]}}`))
-		case 1:
-			conn, _, _ := w.(http.Hijacker).Hijack()
-			_ = conn.Close()
-		case 2:
-			w.Write([]byte(`{"status":"success","data":{"resultType":"scalar","result":[1583786142, "2"]}}`))
-		case 3:
-			conn, _, _ := w.(http.Hijacker).Hijack()
-			_ = conn.Close()
-		case 4:
-			conn, _, _ := w.(http.Hijacker).Hijack()
-			_ = conn.Close()
-		}
-	})
-
-	srv := httptest.NewServer(mux)
-	defer srv.Close()
-
-	s := NewPrometheusClient(srv.URL, nil, false, srv.Client())
-	pq := s.BuildWithParams(QuerierParams{DataSourceType: string(datasourcePrometheus)})
-
-	expErr := func(err string) {
-		_, _, gotErr := pq.Query(ctx, vmQuery, time.Now())
-		if gotErr == nil {
-			t.Fatalf("expected %q got nil", err)
-		}
-		if !strings.Contains(gotErr.Error(), err) {
-			t.Fatalf("expected err %q; got %q", err, gotErr)
-		}
-	}
-
-	expValue := func(v float64) {
-		res, _, err := pq.Query(ctx, vmQuery, time.Now())
-		if err != nil {
-			t.Fatalf("unexpected %s", err)
-		}
-		m := res.Data
-		if len(m) != 1 {
-			t.Fatalf("expected 1 metrics got %d in %+v", len(m), m)
-		}
-		expected := []Metric{
-			{
-				Timestamps: []int64{1583786142},
-				Values:     []float64{v},
-			},
-		}
-		if !reflect.DeepEqual(m, expected) {
-			t.Fatalf("unexpected metric %+v want %+v", m, expected)
-		}
-	}
-
-	expValue(1)   // 0
-	expValue(2)   // 1 - fail, 2 - retry
-	expErr("EOF") // 3, 4 - retries
-}
-
-func metricsEqual(t *testing.T, gotM, expectedM []Metric) {
-	for i, exp := range expectedM {
-		got := gotM[i]
-		gotTS, expTS := got.Timestamps, exp.Timestamps
-		if !reflect.DeepEqual(gotTS, expTS) {
-			t.Fatalf("unexpected timestamps %+v want %+v", gotTS, expTS)
-		}
-		gotV, expV := got.Values, exp.Values
-		if !reflect.DeepEqual(gotV, expV) {
-			t.Fatalf("unexpected values %+v want %+v", gotV, expV)
-		}
-		sort.Slice(got.Labels, func(i, j int) bool {
-			return got.Labels[i].Name < got.Labels[j].Name
-		})
-		sort.Slice(exp.Labels, func(i, j int) bool {
-			return exp.Labels[i].Name < exp.Labels[j].Name
-		})
-		if !reflect.DeepEqual(exp.Labels, got.Labels) {
-			t.Fatalf("unexpected labels %+v want %+v", got.Labels, exp.Labels)
-		}
-	}
-}
-
-func TestVMRangeQuery(t *testing.T) {
-	mux := http.NewServeMux()
-	mux.HandleFunc("/", func(_ http.ResponseWriter, _ *http.Request) {
-		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.Fatalf("expected POST method got %s", r.Method)
-		}
-		if name, pass, _ := r.BasicAuth(); name != basicAuthName || pass != basicAuthPass {
-			t.Fatalf("expected %s:%s as basic auth got %s:%s", basicAuthName, basicAuthPass, name, pass)
-		}
-		if r.URL.Query().Get("query") != vmQuery {
-			t.Fatalf("expected %s in query param, got %s", vmQuery, r.URL.Query().Get("query"))
-		}
-		startTS := r.URL.Query().Get("start")
-		if startTS == "" {
-			t.Fatalf("expected 'start' in query param, got nil instead")
-		}
-		if _, err := time.Parse(time.RFC3339, startTS); err != nil {
-			t.Fatalf("failed to parse 'start' query param: %s", err)
-		}
-		endTS := r.URL.Query().Get("end")
-		if endTS == "" {
-			t.Fatalf("expected 'end' in query param, got nil instead")
-		}
-		if _, err := time.Parse(time.RFC3339, endTS); err != nil {
-			t.Fatalf("failed to parse 'end' query param: %s", err)
-		}
-		step := r.URL.Query().Get("step")
-		if step != "15s" {
-			t.Fatalf("expected 'step' query param to be 15s; got %q instead", step)
-		}
-		switch c {
-		case 0:
-			w.Write([]byte(`{"status":"success","data":{"resultType":"matrix","result":[{"metric":{"__name__":"vm_rows"},"values":[[1583786142,"13763"]]}]}}`))
-		}
-	})
-	mux.HandleFunc("/select/logsql/stats_query_range", func(w http.ResponseWriter, r *http.Request) {
-		c++
-		if r.Method != http.MethodPost {
-			t.Fatalf("expected POST method got %s", r.Method)
-		}
-		if name, pass, _ := r.BasicAuth(); name != basicAuthName || pass != basicAuthPass {
-			t.Fatalf("expected %s:%s as basic auth got %s:%s", basicAuthName, basicAuthPass, name, pass)
-		}
-		if r.URL.Query().Get("query") != vlogsRangeQuery {
-			t.Fatalf("expected %s in query param, got %s", vmQuery, r.URL.Query().Get("query"))
-		}
-		startTS := r.URL.Query().Get("start")
-		if startTS == "" {
-			t.Fatalf("expected 'start' in query param, got nil instead")
-		}
-		if _, err := time.Parse(time.RFC3339, startTS); err != nil {
-			t.Fatalf("failed to parse 'start' query param: %s", err)
-		}
-		endTS := r.URL.Query().Get("end")
-		if endTS == "" {
-			t.Fatalf("expected 'end' in query param, got nil instead")
-		}
-		if _, err := time.Parse(time.RFC3339, endTS); err != nil {
-			t.Fatalf("failed to parse 'end' query param: %s", err)
-		}
-		step := r.URL.Query().Get("step")
-		if step != "60s" {
-			t.Fatalf("expected 'step' query param to be 60s; got %q instead", step)
-		}
-		switch c {
-		case 1:
-			w.Write([]byte(`{"status":"success","data":{"resultType":"matrix","result":[{"metric":{"__name__":"total"},"values":[[1583786142,"10"]]}]}}`))
-		}
-	})
-
-	srv := httptest.NewServer(mux)
-	defer srv.Close()
-
-	authCfg, err := baCfg.NewConfig(".")
-	if err != nil {
-		t.Fatalf("unexpected: %s", err)
-	}
-	s := NewPrometheusClient(srv.URL, authCfg, false, srv.Client())
-
-	pq := s.BuildWithParams(QuerierParams{DataSourceType: string(datasourcePrometheus), EvaluationInterval: 15 * time.Second})
-
-	_, err = pq.QueryRange(ctx, vmQuery, time.Now(), time.Time{})
-	expectError(t, err, "is missing")
-
-	_, err = pq.QueryRange(ctx, vmQuery, time.Time{}, time.Now())
-	expectError(t, err, "is missing")
-
-	start, end := time.Now().Add(-time.Minute), time.Now()
-
-	res, err := pq.QueryRange(ctx, vmQuery, start, end)
-	if err != nil {
-		t.Fatalf("unexpected %s", err)
-	}
-	m := res.Data
-	if len(m) != 1 {
-		t.Fatalf("expected 1 metric  got %d in %+v", len(m), m)
-	}
-	expected := Metric{
-		Labels:     []prompb.Label{{Value: "vm_rows", Name: "__name__"}},
-		Timestamps: []int64{1583786142},
-		Values:     []float64{13763},
-	}
-	if !reflect.DeepEqual(m[0], expected) {
-		t.Fatalf("unexpected metric %+v want %+v", m[0], expected)
-	}
-
-	// test unsupported graphite
-	gq := s.BuildWithParams(QuerierParams{DataSourceType: string(datasourceGraphite)})
-
-	_, err = gq.QueryRange(ctx, queryRender, start, end)
-	expectError(t, err, "is not supported")
-
-	// unsupported logsql
-	gq = s.BuildWithParams(QuerierParams{DataSourceType: string(datasourceVLogs), EvaluationInterval: 60 * time.Second})
-
-	res, err = gq.QueryRange(ctx, vlogsRangeQuery, start, end)
-	expectError(t, err, "is not supported")
-
-	// supported logsql
-	gq = s.BuildWithParams(QuerierParams{DataSourceType: string(datasourceVLogs), EvaluationInterval: 60 * time.Second, ApplyIntervalAsTimeFilter: true})
-	res, err = gq.QueryRange(ctx, vlogsRangeQuery, start, end)
-	if err != nil {
-		t.Fatalf("unexpected %s", err)
-	}
-	m = res.Data
-	if len(m) != 1 {
-		t.Fatalf("expected 1 metric  got %d in %+v", len(m), m)
-	}
-	expected = Metric{
-		Labels:     []prompb.Label{{Value: "total", Name: "stats_result"}},
-		Timestamps: []int64{1583786142},
-		Values:     []float64{10},
-	}
-	if !reflect.DeepEqual(m[0], expected) {
-		t.Fatalf("unexpected metric %+v want %+v", m[0], expected)
-	}
-}
-
-func TestRequestParams(t *testing.T) {
-	query := "up"
-	vlogsQuery := "_time: 5m | stats count() total"
-	timestamp := time.Date(2001, 2, 3, 4, 5, 6, 0, time.UTC)
-
-	f := func(isQueryRange bool, c *Client, checkFn func(t *testing.T, r *http.Request)) {
-		t.Helper()
-
-		req, err := c.newRequest(ctx)
-		if err != nil {
-			t.Fatalf("error in newRequest: %s", err)
-		}
-
-		switch c.dataSourceType {
-		case datasourcePrometheus:
-			if isQueryRange {
-				c.setPrometheusRangeReqParams(req, query, timestamp, timestamp)
-			} else {
-				c.setPrometheusInstantReqParams(req, query, timestamp)
-			}
-		case datasourceGraphite:
-			c.setGraphiteReqParams(req, query)
-		case datasourceVLogs:
-			if isQueryRange {
-				c.setVLogsRangeReqParams(req, vlogsQuery, timestamp, timestamp)
-			} else {
-				c.setVLogsInstantReqParams(req, vlogsQuery, timestamp)
-			}
-		}
-
-		checkFn(t, req)
-	}
-
-	authCfg, err := baCfg.NewConfig(".")
-	if err != nil {
-		t.Fatalf("unexpected error: %s", err)
-	}
-	storage := Client{
-		extraParams: url.Values{"round_digits": {"10"}},
-	}
-
-	// prometheus path
-	f(false, &Client{
-		dataSourceType: datasourcePrometheus,
-	}, func(t *testing.T, r *http.Request) {
-		checkEqualString(t, "/api/v1/query", r.URL.Path)
-	})
-
-	// prometheus prefix
-	f(false, &Client{
-		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, &Client{
-		dataSourceType: datasourcePrometheus,
-	}, func(t *testing.T, r *http.Request) {
-		checkEqualString(t, "/api/v1/query_range", r.URL.Path)
-	})
-
-	// prometheus range prefix
-	f(true, &Client{
-		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, &Client{
-		dataSourceType: datasourceGraphite,
-	}, func(t *testing.T, r *http.Request) {
-		checkEqualString(t, graphitePath, r.URL.Path)
-	})
-
-	// graphite prefix
-	f(false, &Client{
-		dataSourceType:   datasourceGraphite,
-		appendTypePrefix: true,
-	}, func(t *testing.T, r *http.Request) {
-		checkEqualString(t, graphitePrefix+graphitePath, r.URL.Path)
-	})
-
-	// default params
-	f(false, &Client{dataSourceType: datasourcePrometheus}, func(t *testing.T, r *http.Request) {
-		exp := url.Values{"query": {query}, "time": {timestamp.Format(time.RFC3339)}}
-		checkEqualString(t, exp.Encode(), r.URL.RawQuery)
-	})
-
-	f(false, &Client{dataSourceType: datasourcePrometheus, applyIntervalAsTimeFilter: true}, 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, &Client{dataSourceType: datasourcePrometheus}, 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, &Client{
-		dataSourceType: datasourcePrometheus,
-		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, &Client{
-		dataSourceType: datasourcePrometheus,
-		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, &Client{
-		dataSourceType:     datasourcePrometheus,
-		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, &Client{
-		dataSourceType: datasourcePrometheus,
-		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, &Client{
-		dataSourceType:     datasourcePrometheus,
-		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, &Client{
-		dataSourceType: datasourcePrometheus,
-		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, &Client{
-		dataSourceType: datasourcePrometheus,
-		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{
-		DataSourceType: string(datasourcePrometheus),
-		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{
-		DataSourceType: string(datasourcePrometheus),
-		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, &Client{
-		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, &Client{
-		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)
-	})
-
-	// test vlogs
-	f(false, &Client{
-		dataSourceType:     datasourceVLogs,
-		evaluationInterval: time.Minute,
-	}, func(t *testing.T, r *http.Request) {
-		exp := url.Values{"query": {vlogsQuery}, "time": {timestamp.Format(time.RFC3339)}}
-		checkEqualString(t, exp.Encode(), r.URL.RawQuery)
-	})
-
-	f(false, &Client{
-		dataSourceType:            datasourceVLogs,
-		evaluationInterval:        time.Minute,
-		applyIntervalAsTimeFilter: true,
-	}, func(t *testing.T, r *http.Request) {
-		ts := timestamp.Format(time.RFC3339)
-		exp := url.Values{"query": {vlogsQuery}, "time": {ts}, "start": {timestamp.Add(-time.Minute).Format(time.RFC3339)}, "end": {ts}}
-		checkEqualString(t, exp.Encode(), r.URL.RawQuery)
-	})
-
-	f(true, &Client{
-		dataSourceType:     datasourceVLogs,
-		evaluationInterval: time.Minute,
-	}, func(t *testing.T, r *http.Request) {
-		ts := timestamp.Format(time.RFC3339)
-		exp := url.Values{"query": {vlogsQuery}, "start": {ts}, "end": {ts}, "step": {"60s"}}
-		checkEqualString(t, exp.Encode(), r.URL.RawQuery)
-	})
-}
-
-func TestHeaders(t *testing.T) {
-	f := func(vmFn func() *Client, 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() *Client {
-		cfg, err := vmalertutil.AuthConfig(vmalertutil.WithBasicAuth("foo", "bar", ""))
-		if err != nil {
-			t.Fatalf("Error get auth config: %s", err)
-		}
-		return NewPrometheusClient("", cfg, false, nil)
-	}, func(t *testing.T, r *http.Request) {
-		u, p, _ := r.BasicAuth()
-		checkEqualString(t, "foo", u)
-		checkEqualString(t, "bar", p)
-	})
-
-	// bearer auth
-	f(func() *Client {
-		cfg, err := vmalertutil.AuthConfig(vmalertutil.WithBearer("foo", ""))
-		if err != nil {
-			t.Fatalf("Error get auth config: %s", err)
-		}
-		return NewPrometheusClient("", cfg, false, nil)
-	}, 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() *Client {
-		c := NewPrometheusClient("", nil, false, nil)
-		c.extraHeaders = []keyValue{
-			{key: "Foo", value: "bar"},
-			{key: "Baz", value: "qux"},
-		}
-		return c
-	}, 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() *Client {
-		cfg, err := vmalertutil.AuthConfig(vmalertutil.WithBasicAuth("foo", "bar", ""))
-		if err != nil {
-			t.Fatalf("Error get auth config: %s", err)
-		}
-		c := NewPrometheusClient("", cfg, false, nil)
-		c.extraHeaders = []keyValue{
-			{key: "Authorization", value: "Basic QWxhZGRpbjpvcGVuIHNlc2FtZQ=="},
-		}
-		return c
-	}, 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.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.Fatalf("expected non-nil error")
-	}
-	if !strings.Contains(err.Error(), exp) {
-		t.Fatalf("expected error %q to contain %q", err, exp)
-	}
-}

app/vmalert/datasource/client_vlogs.go

@@ -1,61 +0,0 @@
-package datasource
-
-import (
-	"fmt"
-	"net/http"
-	"time"
-)
-
-func (c *Client) setVLogsInstantReqParams(r *http.Request, query string, timestamp time.Time) {
-	// there is no type path prefix in victorialogs APIs right now, ignore appendTypePrefix.
-	if !*disablePathAppend {
-		r.URL.Path += "/select/logsql/stats_query"
-	}
-	q := r.URL.Query()
-	// set `time` param explicitly, it will be used as the timestamp of query results.
-	q.Set("time", timestamp.Format(time.RFC3339))
-	// set the `start` and `end` params if applyIntervalAsTimeFilter is enabled(time filter is missing in the rule expr),
-	// so the query will be executed in time range [timestamp - evaluationInterval, timestamp].
-	if c.applyIntervalAsTimeFilter && c.evaluationInterval > 0 {
-		q.Set("start", timestamp.Add(-c.evaluationInterval).Format(time.RFC3339))
-		q.Set("end", timestamp.Format(time.RFC3339))
-	}
-	r.URL.RawQuery = q.Encode()
-	c.setReqParams(r, query)
-}
-
-func (c *Client) setVLogsRangeReqParams(r *http.Request, query string, start, end time.Time) {
-	// there is no type path prefix in victorialogs APIs right now, ignore appendTypePrefix.
-	if !*disablePathAppend {
-		r.URL.Path += "/select/logsql/stats_query_range"
-	}
-	q := r.URL.Query()
-	q.Add("start", start.Format(time.RFC3339))
-	q.Add("end", end.Format(time.RFC3339))
-	// set step as evaluationInterval by default
-	if c.evaluationInterval > 0 {
-		q.Set("step", fmt.Sprintf("%ds", int(c.evaluationInterval.Seconds())))
-	}
-	r.URL.RawQuery = q.Encode()
-	c.setReqParams(r, query)
-}
-
-func parseVLogsResponse(req *http.Request, resp *http.Response) (res Result, err error) {
-	res, err = parsePrometheusResponse(req, resp)
-	if err != nil {
-		return Result{}, err
-	}
-	for i := range res.Data {
-		m := &res.Data[i]
-		for j := range m.Labels {
-			// reserve the stats func result name with a new label `stats_result` instead of dropping it,
-			// since there could be multiple stats results in a single query, for instance:
-			// 	_time:5m | stats quantile(0.5, request_duration_seconds) p50, quantile(0.9, request_duration_seconds) p90
-			if m.Labels[j].Name == "__name__" {
-				m.Labels[j].Name = "stats_result"
-				break
-			}
-		}
-	}
-	return
-}

app/vmalert/datasource/datasource.go

@@ -1,15 +1,10 @@
 package datasource
 
 import (
-	"bytes"
 	"context"
 	"net/http"
 	"net/url"
-	"sort"
-	"strconv"
 	"time"
-
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompb"
 )
 
 // Querier interface wraps Query and QueryRange methods
@@ -34,9 +29,6 @@ type Result struct {
 	// If nil, then this feature is not supported by the datasource.
 	// SeriesFetched is supported by VictoriaMetrics since v1.90.
 	SeriesFetched *int
-	// IsPartial is used by VictoriaMetrics to indicate
-	// whether response data is partial.
-	IsPartial *bool
 }
 
 // QuerierBuilder builds Querier with given params.
@@ -47,20 +39,16 @@ type QuerierBuilder interface {
 
 // QuerierParams params for Querier.
 type QuerierParams struct {
-	DataSourceType string
-	// ApplyIntervalAsTimeFilter is only valid for vlogs datasource.
-	// Set to true if there is no [timeFilter](https://docs.victoriametrics.com/victorialogs/logsql/#time-filter) in the rule expression,
-	// and we will add evaluation interval as an additional timeFilter when querying.
-	ApplyIntervalAsTimeFilter bool
-	EvaluationInterval        time.Duration
-	QueryParams               url.Values
-	Headers                   map[string]string
-	Debug                     bool
+	DataSourceType     string
+	EvaluationInterval time.Duration
+	QueryParams        url.Values
+	Headers            map[string]string
+	Debug              bool
 }
 
 // Metric is the basic entity which should be return by datasource
 type Metric struct {
-	Labels     []prompb.Label
+	Labels     []Label
 	Timestamps []int64
 	Values     []float64
 }
@@ -77,9 +65,22 @@ func (m *Metric) SetLabel(key, value string) {
 	m.AddLabel(key, value)
 }
 
+// SetLabels sets the given map as Metric labels
+func (m *Metric) SetLabels(ls map[string]string) {
+	var i int
+	m.Labels = make([]Label, len(ls))
+	for k, v := range ls {
+		m.Labels[i] = Label{
+			Name:  k,
+			Value: v,
+		}
+		i++
+	}
+}
+
 // AddLabel appends the given label to the label set
 func (m *Metric) AddLabel(key, value string) {
-	m.Labels = append(m.Labels, prompb.Label{Name: key, Value: value})
+	m.Labels = append(m.Labels, Label{Name: key, Value: value})
 }
 
 // DelLabel deletes the given label from the label set
@@ -102,68 +103,8 @@ func (m *Metric) Label(key string) string {
 	return ""
 }
 
-// Labels is collection of Label
-type Labels []prompb.Label
-
-func (ls Labels) Len() int           { return len(ls) }
-func (ls Labels) Swap(i, j int)      { ls[i], ls[j] = ls[j], ls[i] }
-func (ls Labels) Less(i, j int) bool { return ls[i].Name < ls[j].Name }
-
-func (ls Labels) String() string {
-	var b bytes.Buffer
-
-	b.WriteByte('{')
-	for i, l := range ls {
-		if i > 0 {
-			b.WriteByte(',')
-			b.WriteByte(' ')
-		}
-		b.WriteString(l.Name)
-		b.WriteByte('=')
-		b.WriteString(strconv.Quote(l.Value))
-	}
-	b.WriteByte('}')
-	return b.String()
-}
-
-// LabelCompare return negative if a is less than b, return 0 if they are the same
-// eg.
-// a=[]Label{{Name: "a", Value: "1"}},b=[]Label{{Name: "b", Value: "1"}}, return -1
-// a=[]Label{{Name: "a", Value: "2"}},b=[]Label{{Name: "a", Value: "1"}}, return 1
-// a=[]Label{{Name: "a", Value: "1"}},b=[]Label{{Name: "a", Value: "1"}}, return 0
-func LabelCompare(a, b Labels) int {
-	l := len(a)
-	if len(b) < l {
-		l = len(b)
-	}
-
-	for i := 0; i < l; i++ {
-		if a[i].Name != b[i].Name {
-			if a[i].Name < b[i].Name {
-				return -1
-			}
-			return 1
-		}
-		if a[i].Value != b[i].Value {
-			if a[i].Value < b[i].Value {
-				return -1
-			}
-			return 1
-		}
-	}
-	// if all labels so far were in common, the set with fewer labels comes first.
-	return len(a) - len(b)
-}
-
-// ConvertToLabels convert map to Labels
-func ConvertToLabels(m map[string]string) (labelset Labels) {
-	for k, v := range m {
-		labelset = append(labelset, prompb.Label{
-			Name:  k,
-			Value: v,
-		})
-	}
-	// sort label
-	sort.Slice(labelset, func(i, j int) bool { return labelset[i].Name < labelset[j].Name })
-	return
+// Label represents metric's label
+type Label struct {
+	Name  string
+	Value string
 }

app/vmalert/datasource/faker.go

@@ -1,140 +0,0 @@
-package datasource
-
-import (
-	"context"
-	"net/http"
-	"sync"
-	"time"
-)
-
-// FakeQuerier is a mock querier that return predefined results and error message
-type FakeQuerier struct {
-	sync.Mutex
-	metrics   []Metric
-	err       error
-	isPartial *bool
-}
-
-// SetErr sets query error message
-func (fq *FakeQuerier) SetErr(err error) {
-	fq.Lock()
-	fq.err = err
-	fq.Unlock()
-}
-
-// SetPartialResponse marks query response as partial
-func (fq *FakeQuerier) SetPartialResponse(partial bool) {
-	fq.Lock()
-	fq.isPartial = &partial
-	fq.Unlock()
-}
-
-// Reset reset querier's error message and results
-func (fq *FakeQuerier) Reset() {
-	fq.Lock()
-	fq.err = nil
-	fq.metrics = fq.metrics[:0]
-	fq.isPartial = nil
-	fq.Unlock()
-}
-
-// Add appends metrics to querier result metrics
-func (fq *FakeQuerier) Add(metrics ...Metric) {
-	fq.Lock()
-	fq.metrics = append(fq.metrics, metrics...)
-	fq.Unlock()
-}
-
-// BuildWithParams return FakeQuerier itself
-func (fq *FakeQuerier) BuildWithParams(_ QuerierParams) Querier {
-	return fq
-}
-
-// QueryRange performs query
-func (fq *FakeQuerier) QueryRange(ctx context.Context, q string, _, _ time.Time) (Result, error) {
-	req, _, err := fq.Query(ctx, q, time.Now())
-	return req, err
-}
-
-// Query returns metrics restored in querier
-func (fq *FakeQuerier) Query(_ context.Context, _ string, _ time.Time) (Result, *http.Request, error) {
-	fq.Lock()
-	defer fq.Unlock()
-	if fq.err != nil {
-		return Result{}, nil, fq.err
-	}
-	cp := make([]Metric, len(fq.metrics))
-	copy(cp, fq.metrics)
-	req, _ := http.NewRequest(http.MethodPost, "foo.com", nil)
-	return Result{Data: cp, IsPartial: fq.isPartial}, req, nil
-}
-
-// FakeQuerierWithRegistry can store different results for different query expr
-type FakeQuerierWithRegistry struct {
-	sync.Mutex
-	registry map[string][]Metric
-}
-
-// Set stores query result for given key
-func (fqr *FakeQuerierWithRegistry) Set(key string, metrics ...Metric) {
-	fqr.Lock()
-	if fqr.registry == nil {
-		fqr.registry = make(map[string][]Metric)
-	}
-	fqr.registry[key] = metrics
-	fqr.Unlock()
-}
-
-// Reset clean querier's results registry
-func (fqr *FakeQuerierWithRegistry) Reset() {
-	fqr.Lock()
-	fqr.registry = nil
-	fqr.Unlock()
-}
-
-// BuildWithParams returns itself
-func (fqr *FakeQuerierWithRegistry) BuildWithParams(_ QuerierParams) Querier {
-	return fqr
-}
-
-// QueryRange performs query
-func (fqr *FakeQuerierWithRegistry) QueryRange(ctx context.Context, q string, _, _ time.Time) (Result, error) {
-	req, _, err := fqr.Query(ctx, q, time.Now())
-	return req, err
-}
-
-// Query returns metrics restored in querier registry
-func (fqr *FakeQuerierWithRegistry) Query(_ context.Context, expr string, _ time.Time) (Result, *http.Request, error) {
-	fqr.Lock()
-	defer fqr.Unlock()
-
-	req, _ := http.NewRequest(http.MethodPost, "foo.com", nil)
-	metrics, ok := fqr.registry[expr]
-	if !ok {
-		return Result{}, req, nil
-	}
-	cp := make([]Metric, len(metrics))
-	copy(cp, metrics)
-	return Result{Data: cp}, req, nil
-}
-
-// FakeQuerierWithDelay mock querier with given delay duration
-type FakeQuerierWithDelay struct {
-	FakeQuerier
-	Delay time.Duration
-}
-
-// Query returns query result after delay duration
-func (fqd *FakeQuerierWithDelay) Query(ctx context.Context, expr string, ts time.Time) (Result, *http.Request, error) {
-	timer := time.NewTimer(fqd.Delay)
-	select {
-	case <-ctx.Done():
-	case <-timer.C:
-	}
-	return fqd.FakeQuerier.Query(ctx, expr, ts)
-}
-
-// BuildWithParams returns itself
-func (fqd *FakeQuerierWithDelay) BuildWithParams(_ QuerierParams) Querier {
-	return fqd
-}

app/vmalert/datasource/init.go

@@ -8,18 +8,15 @@ import (
 	"strings"
 	"time"
 
-	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/vmalertutil"
+	"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/utils"
 	"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputil"
-	"github.com/VictoriaMetrics/VictoriaMetrics/lib/promauth"
 )
 
 var (
-	addr = flag.String("datasource.url", "", "Datasource compatible with Prometheus HTTP API. It can be single node VictoriaMetrics or vmselect endpoint. Required parameter. "+
-		"Supports address in the form of IP address with a port (e.g., http://127.0.0.1:8428) or DNS SRV record. "+
-		"See also -remoteRead.disablePathAppend and -datasource.showURL")
+	addr = flag.String("datasource.url", "", "Datasource compatible with Prometheus HTTP API. It can be single node VictoriaMetrics or vmselect URL. Required parameter. "+
+		"E.g. http://127.0.0.1:8428 . See also -remoteRead.disablePathAppend and -datasource.showURL")
 	appendTypePrefix  = flag.Bool("datasource.appendTypePrefix", false, "Whether to add type prefix to -datasource.url based on the query type. Set to true if sending different query types to the vmselect URL.")
-	showDatasourceURL = flag.Bool("datasource.showURL", false, "Whether to avoid stripping sensitive information such as auth headers or passwords from URLs in log messages or UI and exported metrics. "+
+	showDatasourceURL = flag.Bool("datasource.showURL", false, "Whether to show -datasource.url in the exported metrics. "+
 		"It is hidden by default, since it can contain sensitive info such as auth key")
 
 	headers = flag.String("datasource.headers", "", "Optional HTTP extraHeaders to send with each request to the corresponding -datasource.url. "+
@@ -39,23 +36,23 @@ var (
 	tlsCAFile             = flag.String("datasource.tlsCAFile", "", `Optional path to TLS CA file to use for verifying connections to -datasource.url. By default, system CA is used`)
 	tlsServerName         = flag.String("datasource.tlsServerName", "", `Optional TLS server name to use for connections to -datasource.url. By default, the server name from -datasource.url is used`)
 
-	oauth2ClientID         = flag.String("datasource.oauth2.clientID", "", "Optional OAuth2 clientID to use for -datasource.url")
-	oauth2ClientSecret     = flag.String("datasource.oauth2.clientSecret", "", "Optional OAuth2 clientSecret to use for -datasource.url")
-	oauth2ClientSecretFile = flag.String("datasource.oauth2.clientSecretFile", "", "Optional OAuth2 clientSecretFile to use for -datasource.url")
-	oauth2EndpointParams   = flag.String("datasource.oauth2.endpointParams", "", "Optional OAuth2 endpoint parameters to use for -datasource.url . "+
-		`The endpoint parameters must be set in JSON format: {"param1":"value1",...,"paramN":"valueN"}`)
-	oauth2TokenURL = flag.String("datasource.oauth2.tokenUrl", "", "Optional OAuth2 tokenURL to use for -datasource.url")
-	oauth2Scopes   = flag.String("datasource.oauth2.scopes", "", "Optional OAuth2 scopes to use for -datasource.url. Scopes must be delimited by ';'")
+	oauth2ClientID         = flag.String("datasource.oauth2.clientID", "", "Optional OAuth2 clientID to use for -datasource.url. ")
+	oauth2ClientSecret     = flag.String("datasource.oauth2.clientSecret", "", "Optional OAuth2 clientSecret to use for -datasource.url.")
+	oauth2ClientSecretFile = flag.String("datasource.oauth2.clientSecretFile", "", "Optional OAuth2 clientSecretFile to use for -datasource.url. ")
+	oauth2TokenURL         = flag.String("datasource.oauth2.tokenUrl", "", "Optional OAuth2 tokenURL to use for -datasource.url.")
+	oauth2Scopes           = flag.String("datasource.oauth2.scopes", "", "Optional OAuth2 scopes to use for -datasource.url. Scopes must be delimited by ';'")
 
-	queryStep = flag.Duration("datasource.queryStep", 5*time.Minute, "How far a value can fallback to when evaluating queries to the configured -datasource.url and -remoteRead.url. Only valid for prometheus datasource. "+
+	lookBack  = flag.Duration("datasource.lookback", 0, `Lookback defines how far into the past to look when evaluating queries. For example, if the datasource.lookback=5m then param "time" with value now()-5m will be added to every query.`)
+	queryStep = flag.Duration("datasource.queryStep", 5*time.Minute, "How far a value can fallback to when evaluating queries. "+
 		"For example, if -datasource.queryStep=15s then param \"step\" with value \"15s\" will be added to every query. "+
 		"If set to 0, rule's evaluation interval will be used instead.")
-	maxIdleConnections    = flag.Int("datasource.maxIdleConnections", 100, `Defines the number of idle (keep-alive connections) to each configured datasource. Consider setting this value equal to the value: groups_total * group.concurrency. Too low a value may result in a high number of sockets in TIME_WAIT state.`)
-	idleConnectionTimeout = flag.Duration("datasource.idleConnTimeout", 50*time.Second, `Defines a duration for idle (keep-alive connections) to exist. Consider setting this value less than "-http.idleConnTimeout". It must prevent possible "write: broken pipe" and "read: connection reset by peer" errors.`)
-	disableKeepAlive      = flag.Bool("datasource.disableKeepAlive", false, `Whether to disable long-lived connections to the datasource. `+
-		`If true, disables HTTP keep-alive and will only use the connection to the server for a single HTTP request.`)
-	roundDigits = flag.Int("datasource.roundDigits", 0, `Adds "round_digits" GET param to datasource requests which limits the number of digits after the decimal point in response values. `+
-		`Only valid for VictoriaMetrics as the datasource.`)
+	queryTimeAlignment = flag.Bool("datasource.queryTimeAlignment", true, `Whether to align "time" parameter with evaluation interval.`+
+		"Alignment supposed to produce deterministic results despite number of vmalert replicas or time they were started. See more details here https://github.com/VictoriaMetrics/VictoriaMetrics/pull/1257")
+	maxIdleConnections = flag.Int("datasource.maxIdleConnections", 100, `Defines the number of idle (keep-alive connections) to each configured datasource. Consider setting this value equal to the value: groups_total * group.concurrency. Too low a value may result in a high number of sockets in TIME_WAIT state.`)
+	disableKeepAlive   = flag.Bool("datasource.disableKeepAlive", false, `Whether to disable long-lived connections to the datasource. `+
+		`If true, disables HTTP keep-alives and will only use the connection to the server for a single HTTP request.`)
+	roundDigits = flag.Int("datasource.roundDigits", 0, `Adds "round_digits" GET param to datasource requests. `+
+		`In VM "round_digits" limits the number of digits after the decimal point in response values.`)
 )
 
 // InitSecretFlags must be called after flag.Parse and before any logging
@@ -65,11 +62,6 @@ func InitSecretFlags() {
 	}
 }
 
-// ShowDatasourceURL whether to show -datasource.url with sensitive information
-func ShowDatasourceURL() bool {
-	return *showDatasourceURL
-}
-
 // Param represents an HTTP GET param
 type Param struct {
 	Key, Value string
@@ -79,19 +71,19 @@ type Param struct {
 // Provided extraParams will be added as GET params for
 // each request.
 func Init(extraParams url.Values) (QuerierBuilder, error) {
-	if err := httputil.CheckURL(*addr); err != nil {
-		return nil, fmt.Errorf("invalid -datasource.url: %w", err)
+	if *addr == "" {
+		return nil, fmt.Errorf("datasource.url is empty")
 	}
-	tr, err := promauth.NewTLSTransport(*tlsCertFile, *tlsKeyFile, *tlsCAFile, *tlsServerName, *tlsInsecureSkipVerify, "vmalert_datasource")
+
+	tr, err := utils.Transport(*addr, *tlsCertFile, *tlsKeyFile, *tlsCAFile, *tlsServerName, *tlsInsecureSkipVerify)
 	if err != nil {
-		return nil, fmt.Errorf("failed to create transport for -datasource.url=%q: %w", *addr, err)
+		return nil, fmt.Errorf("failed to create transport: %w", err)
 	}
 	tr.DisableKeepAlives = *disableKeepAlive
 	tr.MaxIdleConnsPerHost = *maxIdleConnections
 	if tr.MaxIdleConns != 0 && tr.MaxIdleConns < tr.MaxIdleConnsPerHost {
 		tr.MaxIdleConns = tr.MaxIdleConnsPerHost
 	}
-	tr.IdleConnTimeout = *idleConnectionTimeout
 
 	if extraParams == nil {
 		extraParams = url.Values{}
@@ -100,29 +92,23 @@ func Init(extraParams url.Values) (QuerierBuilder, error) {
 		extraParams.Set("round_digits", fmt.Sprintf("%d", *roundDigits))
 	}
 
-	endpointParams, err := flagutil.ParseJSONMap(*oauth2EndpointParams)
-	if err != nil {
-		return nil, fmt.Errorf("cannot parse JSON for -datasource.oauth2.endpointParams=%s: %w", *oauth2EndpointParams, err)
-	}
-	authCfg, err := vmalertutil.AuthConfig(
-		vmalertutil.WithBasicAuth(*basicAuthUsername, *basicAuthPassword, *basicAuthPasswordFile),
-		vmalertutil.WithBearer(*bearerToken, *bearerTokenFile),
-		vmalertutil.WithOAuth(*oauth2ClientID, *oauth2ClientSecret, *oauth2ClientSecretFile, *oauth2TokenURL, *oauth2Scopes, endpointParams),
-		vmalertutil.WithHeaders(*headers))
+	authCfg, err := utils.AuthConfig(
+		utils.WithBasicAuth(*basicAuthUsername, *basicAuthPassword, *basicAuthPasswordFile),
+		utils.WithBearer(*bearerToken, *bearerTokenFile),
+		utils.WithOAuth(*oauth2ClientID, *oauth2ClientSecret, *oauth2ClientSecretFile, *oauth2TokenURL, *oauth2Scopes),
+		utils.WithHeaders(*headers))
 	if err != nil {
 		return nil, fmt.Errorf("failed to configure auth: %w", err)
 	}
-	_, err = authCfg.GetAuthHeader()
-	if err != nil {
-		return nil, fmt.Errorf("failed to set request auth header to datasource %q: %w", *addr, err)
-	}
 
-	return &Client{
+	return &VMStorage{
 		c:                &http.Client{Transport: tr},
 		authCfg:          authCfg,
 		datasourceURL:    strings.TrimSuffix(*addr, "/"),
 		appendTypePrefix: *appendTypePrefix,
+		lookBack:         *lookBack,
 		queryStep:        *queryStep,
+		dataSourceType:   datasourcePrometheus,
 		extraParams:      extraParams,
 	}, nil
 }