mudhorn/prometheus
1
0
Fork 0
mirror of https://github.com/prometheus/prometheus.git synced 2025-03-05 20:59:13 -08:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

Merge branch 'main' of github.com:prometheus/prometheus into document-and-test-relabel-memory-reuse

Browse source 
Signed-off-by: Douglas Camata <159076+douglascamata@users.noreply.github.com>
This commit is contained in:
Douglas Camata 2023-07-12 15:01:29 +02:00
parent 28a66e183d 162612ea86
commit 47819fef01
No known key found for this signature in database
428 changed files with 54792 additions and 9903 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
.github/CODEOWNERS vendored
View file

@ -2,6 +2,6 @@
/web/ui/module @juliusv @nexucis
/storage/remote @csmarchbanks @cstyan @bwplotka @tomwilkie
/discovery/kubernetes @brancz
/tsdb @codesome
/promql @codesome @roidelapluie
/tsdb @jesusvazquez
/promql @roidelapluie
/cmd/promtool @dgl

26
.github/ISSUE_TEMPLATE/feature_request.md vendored
View file

@ -1,26 +0,0 @@
---
name: Feature request
about: Suggest an idea for this project.
title: ''
labels: ''
assignees: ''
---
<!--
Please do *NOT* ask support questions in Github issues.
If your issue is not a feature request or bug report use our
community support.
https://prometheus.io/community/
There is also commercial support available.
https://prometheus.io/support-training/
-->
## Proposal
**Use case. Why is this important?**
*“Nice to have” is not a good use case. :)*

23
.github/ISSUE_TEMPLATE/feature_request.yml vendored Normal file
View file

@ -0,0 +1,23 @@
---
name: Feature request
description: Suggest an idea for this project.
body:
- type: markdown
attributes:
value: >-
Please do *NOT* ask support questions in Github issues.
If your issue is not a feature request or bug report use
our [community support](https://prometheus.io/community/).
There is also [commercial
support](https://prometheus.io/support-training/) available.
- type: textarea
attributes:
label: Proposal
description: Use case. Why is this important?
placeholder: “Nice to have” is not a good use case. :)
validations:
required: true

23
.github/actions/build/action.yml vendored
View file

@ -1,23 +0,0 @@
name: Build
inputs:
thread:
type: integer
description: Current thread
required: true
default: 3
parallelism:
type: integer
description: Number of builds to do in parallel
default: 3
promu_opts:
type: string
description: Options to pass to promu
runs:
using: composite
steps:
- uses: ./.github/actions/setup_environment
- run: ~/go/bin/promu crossbuild -v --parallelism ${{ inputs.parallelism }} --parallelism-thread ${{ inputs.thread }} ${{ inputs.promu_opts }}
shell: bash
- uses: ./.github/actions/save_artifacts
with:
directory: .build

20
.github/actions/check_proto/action.yml vendored
View file

@ -1,20 +0,0 @@
name: Check proto files
inputs:
version:
type: string
description: Protoc version
default: "3.5.1"
runs:
using: composite
steps:
- run: |
env
set -x
curl -s -L https://github.com/protocolbuffers/protobuf/releases/download/v${{ inputs.version }}/protoc-${{ inputs.version }}-linux-x86_64.zip > /tmp/protoc.zip
unzip -d /tmp /tmp/protoc.zip
chmod +x /tmp/bin/protoc
export PATH=/tmp/bin:$PATH
make proto
shell: bash
- run: git diff --exit-code
shell: bash

47
.github/actions/publish_images/action.yml vendored
View file

@ -1,47 +0,0 @@
name: Publish image
inputs:
registry:
type: string
description: Docker registry
organization:
type: string
description: Organization
login:
type: string
description: Username
password:
type: string
description: Password
dockerfile_path:
description: Path to Dockerfile
type: string
default: ""
dockerbuild_context:
description: Path to Dockerbuild context
type: string
default: ""
container_image_name:
description: Name of the container image
type: string
default: ""
runs:
using: composite
steps:
- if: inputs.dockerfile_path != ''
run: echo "export DOCKERFILE_PATH=${{ inputs.dockerfile_path }}" >> /tmp/tmp-profile
shell: bash
- if: inputs.container_image_name != ''
run: echo "export DOCKER_IMAGE_NAME=${{ inputs.container_image_name }}" >> /tmp/tmp-profile
shell: bash
- if: inputs.dockerbuild_context != ''
run: echo "export DOCKERBUILD_CONTEXT=${{ inputs.dockerbuild_context }}" >> /tmp/tmp-profile
shell: bash
- run: |
touch /tmp/tmp-profile
. /tmp/tmp-profile
make docker DOCKER_REPO=${{ inputs.registry }}/${{ inputs.organization }}
docker images
echo ${{ inputs.password }} | docker login -u ${{ inputs.login }} --password-stdin ${{ inputs.registry }}
make docker-publish DOCKER_REPO=${{ inputs.registry }}/${{ inputs.organization }}
make docker-manifest DOCKER_REPO=${{ inputs.registry }}/${{ inputs.organization }}
shell: bash

43
.github/actions/publish_main/action.yml vendored
View file

@ -1,43 +0,0 @@
name: Publish image
inputs:
docker_hub_organization:
type: string
description: DockerHub organization
default: prom
docker_hub_login:
type: string
description: DockerHub username
docker_hub_password:
type: string
description: DockerHub password
quay_io_organization:
type: string
description: Quay.io organization
default: prometheus
quay_io_login:
type: string
description: Quay.io username
quay_io_password:
type: string
description: Quay.io password
runs:
using: composite
steps:
- uses: ./.github/actions/setup_environment
with:
enable_docker_multibuild: true
- uses: ./.github/actions/restore_artifacts
- uses: ./.github/actions/publish_images
if: inputs.docker_hub_organization != '' && inputs.docker_hub_login != ''
with:
registry: docker.io
organization: ${{ inputs.docker_hub_organization }}
login: ${{ inputs.docker_hub_login }}
password: ${{ inputs.docker_hub_password }}
- uses: ./.github/actions/publish_images
if: inputs.quay_io_organization != '' && inputs.quay_io_login != ''
with:
registry: quay.io
organization: ${{ inputs.quay_io_organization }}
login: ${{ inputs.quay_io_login }}
password: ${{ inputs.quay_io_password }}

54
.github/actions/publish_release/action.yml vendored
View file

@ -1,54 +0,0 @@
name: Publish image
inputs:
docker_hub_organization:
type: string
description: DockerHub organization
default: prom
docker_hub_login:
type: string
description: DockerHub username
docker_hub_password:
type: string
description: DockerHub password
quay_io_organization:
type: string
description: Quay.io organization
default: prometheus
quay_io_login:
type: string
description: Quay.io username
quay_io_password:
type: string
description: Quay.io password
github_token:
type: string
description: Github Token
runs:
using: composite
steps:
- uses: ./.github/actions/setup_environment
with:
enable_docker_multibuild: true
- uses: ./.github/actions/restore_artifacts
- run: ~/go/bin/promu crossbuild tarballs
shell: bash
- run: ~/go/bin/promu checksum .tarballs
shell: bash
- run: ~/go/bin/promu release .tarballs
shell: bash
env:
GITHUB_TOKEN: ${{ inputs.github_token }}
- uses: ./.github/actions/publish_release_images
if: inputs.docker_hub_organization != '' && inputs.docker_hub_login != ''
with:
registry: docker.io
organization: ${{ inputs.docker_hub_organization }}
login: ${{ inputs.docker_hub_login }}
password: ${{ inputs.docker_hub_password }}
- uses: ./.github/actions/publish_release_images
if: inputs.quay_io_organization != '' && inputs.quay_io_login != ''
with:
registry: quay.io
organization: ${{ inputs.quay_io_organization }}
login: ${{ inputs.quay_io_login }}
password: ${{ inputs.quay_io_password }}

53
.github/actions/publish_release_images/action.yml vendored
View file

@ -1,53 +0,0 @@
name: Publish release image
inputs:
registry:
type: string
description: Docker registry
organization:
type: string
description: Organization
login:
type: string
description: Username
password:
type: string
description: Password
dockerfile_path:
description: Path to Dockerfile
type: string
default: ""
dockerbuild_context:
description: Path to Dockerbuild context
type: string
default: ""
container_image_name:
description: Name of the container image
type: string
default: ""
runs:
using: composite
steps:
- if: inputs.dockerfile_path != ''
run: echo "export DOCKERFILE_PATH=${{ inputs.dockerfile_path }}" >> /tmp/tmp-profile
shell: bash
- if: inputs.container_image_name != ''
run: echo "export DOCKER_IMAGE_NAME=${{ inputs.container_image_name }}" >> /tmp/tmp-profile
shell: bash
- if: inputs.dockerbuild_context != ''
run: echo "export DOCKERBUILD_CONTEXT=${{ inputs.dockerbuild_context }}" >> /tmp/tmp-profile
shell: bash
- run: |
current_tag=${GITHUB_REF#refs/*/}
touch /tmp/tmp-profile
. /tmp/tmp-profile
make docker DOCKER_IMAGE_TAG="$current_tag" DOCKER_REPO=${{ inputs.registry }}/${{ inputs.organization }}
docker images
echo ${{ inputs.password }} | docker login -u ${{ inputs.login }} --password-stdin ${{ inputs.registry }}
make docker-publish DOCKER_IMAGE_TAG="$current_tag" DOCKER_REPO=${{ inputs.registry }}/${{ inputs.organization }}
make docker-manifest DOCKER_IMAGE_TAG="$current_tag" DOCKER_REPO=${{ inputs.registry }}/${{ inputs.organization }}
if [[ "$current_tag" =~ ^v[0-9]+(\.[0-9]+){2}$ ]]; then
make docker-tag-latest DOCKER_IMAGE_TAG="$current_tag" DOCKER_REPO=${{ inputs.registry }}/${{ inputs.organization }}
make docker-publish DOCKER_IMAGE_TAG="latest" DOCKER_REPO=${{ inputs.registry }}/${{ inputs.organization }}
make docker-manifest DOCKER_IMAGE_TAG="latest" DOCKER_REPO=${{ inputs.registry }}/${{ inputs.organization }}
fi
shell: bash

19
.github/actions/restore_artifacts/action.yml vendored
View file

@ -1,19 +0,0 @@
# Restore artifacts created by save_artifacts.
# Tar is used because the default actions do not preserve directory structure
# and file mode.
name: Restore artifacts
runs:
using: composite
steps:
- name: Download all workflow run artifacts
uses: actions/download-artifact@v3
with:
name: artifact
path: .artifacts
- run: |
for tar in .artifacts/*.tar
do
tar xvf $tar
done
rm -v .artifacts/*.tar
shell: bash

17
.github/actions/save_artifacts/action.yml vendored
View file

@ -1,17 +0,0 @@
# Tar is used because the default actions do not preserve directory structure
# and file mode.
name: Save artifacts
inputs:
directory:
type: string
description: Path of the directory to save
runs:
using: composite
steps:
- run: |
tar cvf artifact.tar ${{ inputs.directory }}
mv artifact.tar artifact-$(sha1sum artifact.tar|awk '{ print $1 }').tar
shell: bash
- uses: actions/upload-artifact@v3
with:
path: artifact-*.tar

43
.github/actions/setup_environment/action.yml vendored
View file

@ -1,43 +0,0 @@
name: Setup environment
inputs:
enable_go:
type: boolean
description: Whether to enable go specific features, such as caching.
default: true
enable_npm:
type: boolean
description: Whether to enable npm specific features, such as caching.
default: false
enable_docker_multibuild:
type: boolean
description: Whether to enable multibuild docker
default: false
runs:
using: composite
steps:
- uses: actions/cache@v3
if: inputs.enable_go == 'true'
with:
path: |
~/.cache/go-build
~/go/pkg/mod
key: ${{ runner.os }}-go-${{ hashFiles('**/go.sum') }}
restore-keys: |
${{ runner.os }}-go-
- uses: actions/cache@v3
if: inputs.enable_npm == 'true'
with:
path: |
~/.npm
key: ${{ runner.os }}-npm-${{ hashFiles('web/ui/package-lock.json') }}
restore-keys: |
${{ runner.os }}-npm-
- run: make promu
shell: bash
if: inputs.enable_go == 'true'
- name: Set up QEMU
uses: docker/setup-qemu-action@v1
if: inputs.enable_docker_multibuild == 'true'
- name: Set up buildx
uses: docker/setup-buildx-action@v1
if: inputs.enable_docker_multibuild == 'true'

6
.github/workflows/buf-lint.yml vendored
View file

@ -1,5 +1,5 @@
name: buf.build
on: # yamllint disable-line rule:truthy
on:
pull_request:
paths:
- ".github/workflows/buf-lint.yml"
@ -10,7 +10,9 @@ jobs:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- uses: bufbuild/buf-setup-action@v1.7.0
- uses: bufbuild/buf-setup-action@v1.23.1
with:
github_token: ${{ secrets.GITHUB_TOKEN }}
- uses: bufbuild/buf-lint-action@v1
with:
input: 'prompb'

7
.github/workflows/buf.yml vendored
View file

@ -1,5 +1,5 @@
name: buf.build
on: # yamllint disable-line rule:truthy
on:
push:
branches:
- main
@ -7,9 +7,12 @@ jobs:
buf:
name: lint and publish
runs-on: ubuntu-latest
if: github.repository_owner == 'prometheus'
steps:
- uses: actions/checkout@v3
- uses: bufbuild/buf-setup-action@v1.7.0
- uses: bufbuild/buf-setup-action@v1.23.1
with:
github_token: ${{ secrets.GITHUB_TOKEN }}
- uses: bufbuild/buf-lint-action@v1
with:
input: 'prompb'

49
.github/workflows/ci.yml vendored
View file

@ -1,8 +1,9 @@
---
name: CI
on: # yamllint disable-line rule:truthy
on:
pull_request:
push:
jobs:
test_go:
name: Go tests
@ -10,15 +11,18 @@ jobs:
# Whenever the Go version is updated here, .promu.yml
# should also be updated.
container:
image: quay.io/prometheus/golang-builder:1.19-base
image: quay.io/prometheus/golang-builder:1.20-base
steps:
- uses: actions/checkout@v3
- uses: ./.github/actions/setup_environment
- uses: prometheus/promci@v0.1.0
- uses: ./.github/promci/actions/setup_environment
- run: make GO_ONLY=1 SKIP_GOLANGCI_LINT=1
- run: go test ./tsdb/ -test.tsdb-isolation=false
- run: go test --tags=stringlabels ./...
- run: GOARCH=386 go test ./cmd/prometheus
- run: make -C documentation/examples/remote_storage
- run: make -C documentation/examples
- uses: ./.github/actions/check_proto
- uses: ./.github/promci/actions/check_proto
with:
version: "3.15.8"
@ -28,18 +32,19 @@ jobs:
# Whenever the Go version is updated here, .promu.yml
# should also be updated.
container:
image: quay.io/prometheus/golang-builder:1.19-base
image: quay.io/prometheus/golang-builder:1.20-base
steps:
- uses: actions/checkout@v3
- uses: ./.github/actions/setup_environment
- uses: prometheus/promci@v0.1.0
- uses: ./.github/promci/actions/setup_environment
with:
enable_go: false
enable_npm: true
- run: make assets-tarball
- run: make ui-lint
- run: make ui-test
- uses: ./.github/actions/save_artifacts
- uses: ./.github/promci/actions/save_artifacts
with:
directory: .tarballs
@ -48,9 +53,9 @@ jobs:
runs-on: windows-latest
steps:
- uses: actions/checkout@v3
- uses: actions/setup-go@v3
- uses: actions/setup-go@v4
with:
go-version: '>=1.19 <1.20'
go-version: '>=1.20 <1.21'
- run: |
$TestTargets = go list ./... | Where-Object { $_ -NotMatch "(github.com/prometheus/prometheus/discovery.*|github.com/prometheus/prometheus/config|github.com/prometheus/prometheus/web)"}
go test $TestTargets -vet=off -v
@ -61,7 +66,7 @@ jobs:
runs-on: ubuntu-latest
# The go verson in this image should be N-1 wrt test_go.
container:
image: quay.io/prometheus/golang-builder:1.18-base
image: quay.io/prometheus/golang-builder:1.19-base
steps:
- uses: actions/checkout@v3
- run: make build
@ -100,7 +105,8 @@ jobs:
thread: [ 0, 1, 2 ]
steps:
- uses: actions/checkout@v3
- uses: ./.github/actions/build
- uses: prometheus/promci@v0.1.0
- uses: ./.github/promci/actions/build
with:
promu_opts: "-p linux/amd64 -p windows/amd64 -p linux/arm64 -p darwin/amd64 -p darwin/arm64 -p linux/386"
parallelism: 3
@ -122,7 +128,8 @@ jobs:
# should also be updated.
steps:
- uses: actions/checkout@v3
- uses: ./.github/actions/build
- uses: prometheus/promci@v0.1.0
- uses: ./.github/promci/actions/build
with:
parallelism: 12
thread: ${{ matrix.thread }}
@ -133,16 +140,17 @@ jobs:
- name: Checkout repository
uses: actions/checkout@v3
- name: Install Go
uses: actions/setup-go@v3
uses: actions/setup-go@v4
with:
go-version: '<1.19'
go-version: 1.20.x
- name: Install snmp_exporter/generator dependencies
run: sudo apt-get update && sudo apt-get -y install libsnmp-dev
if: github.repository == 'prometheus/snmp_exporter'
- name: Lint
uses: golangci/golangci-lint-action@v3.2.0
uses: golangci/golangci-lint-action@v3.6.0
with:
version: v1.49.0
args: --verbose
version: v1.53.3
fuzzing:
uses: ./.github/workflows/fuzzing.yml
if: github.event_name == 'pull_request'
@ -156,7 +164,8 @@ jobs:
if: github.event_name == 'push' && github.event.ref == 'refs/heads/main'
steps:
- uses: actions/checkout@v3
- uses: ./.github/actions/publish_main
- uses: prometheus/promci@v0.1.0
- uses: ./.github/promci/actions/publish_main
with:
docker_hub_login: ${{ secrets.docker_hub_login }}
docker_hub_password: ${{ secrets.docker_hub_password }}
@ -169,7 +178,8 @@ jobs:
if: github.event_name == 'push' && startsWith(github.ref, 'refs/tags/v2.')
steps:
- uses: actions/checkout@v3
- uses: ./.github/actions/publish_release
- uses: prometheus/promci@v0.1.0
- uses: ./.github/promci/actions/publish_release
with:
docker_hub_login: ${{ secrets.docker_hub_login }}
docker_hub_password: ${{ secrets.docker_hub_password }}
@ -183,12 +193,13 @@ jobs:
steps:
- name: Checkout
uses: actions/checkout@v3
- uses: prometheus/promci@v0.1.0
- name: Install nodejs
uses: actions/setup-node@v3
with:
node-version-file: "web/ui/.nvmrc"
registry-url: "https://registry.npmjs.org"
- uses: actions/cache@v3.0.9
- uses: actions/cache@v3.3.1
with:
path: ~/.npm
key: ${{ runner.os }}-node-${{ hashFiles('**/package-lock.json') }}

6
.github/workflows/codeql-analysis.yml vendored
View file

@ -1,7 +1,7 @@
---
name: "CodeQL"
on: # yamllint disable-line rule:truthy
on:
workflow_call:
schedule:
- cron: "26 14 * * 1"
@ -21,9 +21,9 @@ jobs:
steps:
- name: Checkout repository
uses: actions/checkout@v3
- uses: actions/setup-go@v3
- uses: actions/setup-go@v4
with:
go-version: '>=1.19 <1.20'
go-version: '>=1.20 <1.21'
- name: Initialize CodeQL
uses: github/codeql-action/init@v2

2
.github/workflows/funcbench.yml vendored
View file

@ -1,4 +1,4 @@
on: # yamllint disable-line rule:truthy
on:
repository_dispatch:
types: [funcbench_start]
name: Funcbench Workflow

2
.github/workflows/fuzzing.yml vendored
View file

@ -1,5 +1,5 @@
name: CIFuzz
on: # yamllint disable-line rule:truthy
on:
workflow_call:
jobs:
Fuzzing:

5
.github/workflows/lock.yml vendored
View file

@ -1,6 +1,6 @@
name: 'Lock Threads'
on: # yamllint disable-line rule:truthy
on:
schedule:
- cron: '13 23 * * *'
workflow_dispatch:
@ -14,8 +14,9 @@ concurrency:
jobs:
action:
runs-on: ubuntu-latest
if: github.repository_owner == 'prometheus'
steps:
- uses: dessant/lock-threads@v3
- uses: dessant/lock-threads@v4
with:
process-only: 'issues'
issue-inactive-days: '180'

2
.github/workflows/prombench.yml vendored
View file

@ -1,4 +1,4 @@
on: # yamllint disable-line rule:truthy
on:
repository_dispatch:
types: [prombench_start, prombench_restart, prombench_stop]
name: Prombench Workflow

3
.github/workflows/repo_sync.yml vendored
View file

@ -1,11 +1,12 @@
---
name: Sync repo files
on: # yamllint disable-line rule:truthy
on:
schedule:
- cron: '44 17 * * *'
jobs:
repo_sync:
runs-on: ubuntu-latest
if: github.repository_owner == 'prometheus'
container:
image: quay.io/prometheus/golang-builder
steps:

5
.gitignore vendored
View file

@ -19,7 +19,7 @@ benchmark.txt
!/.promu.yml
!/.golangci.yml
/documentation/examples/remote_storage/remote_storage_adapter/remote_storage_adapter
/documentation/examples/remote_storage/example_write_adapter/example_writer_adapter
/documentation/examples/remote_storage/example_write_adapter/example_write_adapter
npm_licenses.tar.bz2
/web/ui/static/react
@ -28,3 +28,6 @@ npm_licenses.tar.bz2
/.build
/**/node_modules
# Ignore parser debug
y.output

51
.golangci.yml
View file

@ -1,5 +1,5 @@
run:
deadline: 5m
timeout: 15m
skip-files:
# Skip autogenerated files.
- ^.*\.(pb|y)\.go$
@ -10,31 +10,60 @@ output:
linters:
enable:
- depguard
- gocritic
- gofumpt
- goimports
- revive
- misspell
- predeclared
- revive
- unconvert
- unused
issues:
max-same-issues: 0
exclude-rules:
- linters:
- gocritic
text: "appendAssign"
- path: _test.go
linters:
- errcheck
linters-settings:
depguard:
list-type: blacklist
include-go-root: true
packages-with-error-message:
- sync/atomic: "Use go.uber.org/atomic instead of sync/atomic"
- github.com/stretchr/testify/assert: "Use github.com/stretchr/testify/require instead of github.com/stretchr/testify/assert"
- github.com/go-kit/kit/log: "Use github.com/go-kit/log instead of github.com/go-kit/kit/log"
- io/ioutil: "Use corresponding 'os' or 'io' functions instead."
- regexp: "Use github.com/grafana/regexp instead of regexp"
rules:
main:
deny:
- pkg: "sync/atomic"
desc: "Use go.uber.org/atomic instead of sync/atomic"
- pkg: "github.com/stretchr/testify/assert"
desc: "Use github.com/stretchr/testify/require instead of github.com/stretchr/testify/assert"
- pkg: "github.com/go-kit/kit/log"
desc: "Use github.com/go-kit/log instead of github.com/go-kit/kit/log"
- pkg: "io/ioutil"
desc: "Use corresponding 'os' or 'io' functions instead."
- pkg: "regexp"
desc: "Use github.com/grafana/regexp instead of regexp"
errcheck:
exclude: scripts/errcheck_excludes.txt
exclude-functions:
# Don't flag lines such as "io.Copy(io.Discard, resp.Body)".
- io.Copy
# The next two are used in HTTP handlers, any error is handled by the server itself.
- io.WriteString
- (net/http.ResponseWriter).Write
# No need to check for errors on server's shutdown.
- (*net/http.Server).Shutdown
# Never check for logger errors.
- (github.com/go-kit/log.Logger).Log
# Never check for rollback errors as Rollback() is called when a previous error was detected.
- (github.com/prometheus/prometheus/storage.Appender).Rollback
goimports:
local-prefixes: github.com/prometheus/prometheus
gofumpt:
extra-rules: true
revive:
rules:
# https://github.com/mgechev/revive/blob/master/RULES_DESCRIPTIONS.md#unused-parameter
- name: unused-parameter
severity: warning
disabled: true

12
.promu.yml
View file

@ -1,7 +1,7 @@
go:
# Whenever the Go version is updated here,
# .circle/config.yml should also be updated.
version: 1.19
version: 1.20
repository:
path: github.com/prometheus/prometheus
build:
@ -10,7 +10,15 @@ build:
path: ./cmd/prometheus
- name: promtool
path: ./cmd/promtool
flags: -a -tags netgo,builtinassets
tags:
all:
- netgo
- builtinassets
- stringlabels
windows:
- builtinassets
- stringlabels
flags: -a
ldflags: |
-X github.com/prometheus/common/version.Version={{.Version}}
-X github.com/prometheus/common/version.Revision={{.Revision}}

7
.yamllint
View file

@ -20,9 +20,4 @@ rules:
config/testdata/section_key_dup.bad.yml
line-length: disable
truthy:
ignore: |
.github/workflows/codeql-analysis.yml
.github/workflows/funcbench.yml
.github/workflows/fuzzing.yml
.github/workflows/prombench.yml
.github/workflows/golangci-lint.yml
check-keys: false

168
CHANGELOG.md
View file

@ -1,5 +1,173 @@
# Changelog
## 2.45.0 / 2023-06-23
This release is a LTS (Long-Term Support) release of Prometheus and will
receive security, documentation and bugfix patches for at least 12 months.
Please read more about our LTS release cycle at
<https://prometheus.io/docs/introduction/release-cycle/>.
* [FEATURE] API: New limit parameter to limit the number of items returned by `/api/v1/status/tsdb` endpoint. #12336
* [FEATURE] Config: Add limits to global config. #12126
* [FEATURE] Consul SD: Added support for `path_prefix`. #12372
* [FEATURE] Native histograms: Add option to scrape both classic and native histograms. #12350
* [FEATURE] Native histograms: Added support for two more arithmetic operators `avg_over_time` and `sum_over_time`. #12262
* [FEATURE] Promtool: When providing the block id, only one block will be loaded and analyzed. #12031
* [FEATURE] Remote-write: New Azure ad configuration to support remote writing directly to Azure Monitor workspace. #11944
* [FEATURE] TSDB: Samples per chunk are now configurable with flag `storage.tsdb.samples-per-chunk`. By default set to its former value 120. #12055
* [ENHANCEMENT] Native histograms: bucket size can now be limited to avoid scrape fails. #12254
* [ENHANCEMENT] TSDB: Dropped series are now deleted from the WAL sooner. #12297
* [BUGFIX] Native histograms: ChunkSeries iterator now checks if a new sample can be appended to the open chunk. #12185
* [BUGFIX] Native histograms: Fix Histogram Appender `Appendable()` segfault. #12357
* [BUGFIX] Native histograms: Fix setting reset header to gauge histograms in seriesToChunkEncoder. #12329
* [BUGFIX] TSDB: Tombstone intervals are not modified after Get() call. #12245
* [BUGFIX] TSDB: Use path/filepath to set the WAL directory. #12349
## 2.44.0 / 2023-05-13
This version is built with Go tag `stringlabels`, to use the smaller data
structure for Labels that was optional in the previous release. For more
details about this code change see #10991.
* [CHANGE] Remote-write: Raise default samples per send to 2,000. #12203
* [FEATURE] Remote-read: Handle native histograms. #12085, #12192
* [FEATURE] Promtool: Health and readiness check of prometheus server in CLI. #12096
* [FEATURE] PromQL: Add `query_samples_total` metric, the total number of samples loaded by all queries. #12251
* [ENHANCEMENT] Storage: Optimise buffer used to iterate through samples. #12326
* [ENHANCEMENT] Scrape: Reduce memory allocations on target labels. #12084
* [ENHANCEMENT] PromQL: Use faster heap method for `topk()` / `bottomk()`. #12190
* [ENHANCEMENT] Rules API: Allow filtering by rule name. #12270
* [ENHANCEMENT] Native Histograms: Various fixes and improvements. #11687, #12264, #12272
* [ENHANCEMENT] UI: Search of scraping pools is now case-insensitive. #12207
* [ENHANCEMENT] TSDB: Add an affirmative log message for successful WAL repair. #12135
* [BUGFIX] TSDB: Block compaction failed when shutting down. #12179
* [BUGFIX] TSDB: Out-of-order chunks could be ignored if the write-behind log was deleted. #12127
## 2.43.1 / 2023-05-03
* [BUGFIX] Labels: `Set()` after `Del()` would be ignored, which broke some relabeling rules. #12322
## 2.43.0 / 2023-03-21
We are working on some performance improvements in Prometheus, which are only
built into Prometheus when compiling it using the Go tag `stringlabels`
(therefore they are not shipped in the default binaries). It uses a data
structure for labels that uses a single string to hold all the label/values,
resulting in a smaller heap size and some speedups in most cases. We would like
to encourage users who are interested in these improvements to help us measure
the gains on their production architecture. We are providing release artefacts
`2.43.0+stringlabels` and Docker images tagged `v2.43.0-stringlabels` with those
improvements for testing. #10991
* [FEATURE] Promtool: Add HTTP client configuration to query commands. #11487
* [FEATURE] Scrape: Add `scrape_config_files` to include scrape configs from different files. #12019
* [FEATURE] HTTP client: Add `no_proxy` to exclude URLs from proxied requests. #12098
* [FEATURE] HTTP client: Add `proxy_from_environment` to read proxies from env variables. #12098
* [ENHANCEMENT] API: Add support for setting lookback delta per query via the API. #12088
* [ENHANCEMENT] API: Change HTTP status code from 503/422 to 499 if a request is canceled. #11897
* [ENHANCEMENT] Scrape: Allow exemplars for all metric types. #11984
* [ENHANCEMENT] TSDB: Add metrics for head chunks and WAL folders size. #12013
* [ENHANCEMENT] TSDB: Automatically remove incorrect snapshot with index that is ahead of WAL. #11859
* [ENHANCEMENT] TSDB: Improve Prometheus parser error outputs to be more comprehensible. #11682
* [ENHANCEMENT] UI: Scope `group by` labels to metric in autocompletion. #11914
* [BUGFIX] Scrape: Fix `prometheus_target_scrape_pool_target_limit` metric not set before reloading. #12002
* [BUGFIX] TSDB: Correctly update `prometheus_tsdb_head_chunks_removed_total` and `prometheus_tsdb_head_chunks` metrics when reading WAL. #11858
* [BUGFIX] TSDB: Use the correct unit (seconds) when recording out-of-order append deltas in the `prometheus_tsdb_sample_ooo_delta` metric. #12004
## 2.42.0 / 2023-01-31
This release comes with a bunch of feature coverage for native histograms and breaking changes.
If you are trying native histograms already, we recommend you remove the `wal` directory when upgrading.
Because the old WAL record for native histograms is not backward compatible in v2.42.0, this will lead to some data loss for the latest data.
Additionally, if you scrape "float histograms" or use recording rules on native histograms in v2.42.0 (which writes float histograms),
it is a one-way street since older versions do not support float histograms.
* [CHANGE] **breaking** TSDB: Changed WAL record format for the experimental native histograms. #11783
* [FEATURE] Add 'keep_firing_for' field to alerting rules. #11827
* [FEATURE] Promtool: Add support of selecting timeseries for TSDB dump. #11872
* [ENHANCEMENT] Agent: Native histogram support. #11842
* [ENHANCEMENT] Rules: Support native histograms in recording rules. #11838
* [ENHANCEMENT] SD: Add container ID as a meta label for pod targets for Kubernetes. #11844
* [ENHANCEMENT] SD: Add VM size label to azure service discovery. #11650
* [ENHANCEMENT] Support native histograms in federation. #11830
* [ENHANCEMENT] TSDB: Add gauge histogram support. #11783 #11840 #11814
* [ENHANCEMENT] TSDB/Scrape: Support FloatHistogram that represents buckets as float64 values. #11522 #11817 #11716
* [ENHANCEMENT] UI: Show individual scrape pools on /targets page. #11142
## 2.41.0 / 2022-12-20
* [FEATURE] Relabeling: Add `keepequal` and `dropequal` relabel actions. #11564
* [FEATURE] Add support for HTTP proxy headers. #11712
* [ENHANCEMENT] Reload private certificates when changed on disk. #11685
* [ENHANCEMENT] Add `max_version` to specify maximum TLS version in `tls_config`. #11685
* [ENHANCEMENT] Add `goos` and `goarch` labels to `prometheus_build_info`. #11685
* [ENHANCEMENT] SD: Add proxy support for EC2 and LightSail SDs #11611
* [ENHANCEMENT] SD: Add new metric `prometheus_sd_file_watcher_errors_total`. #11066
* [ENHANCEMENT] Remote Read: Use a pool to speed up marshalling. #11357
* [ENHANCEMENT] TSDB: Improve handling of tombstoned chunks in iterators. #11632
* [ENHANCEMENT] TSDB: Optimize postings offset table reading. #11535
* [BUGFIX] Scrape: Validate the metric name, label names, and label values after relabeling. #11074
* [BUGFIX] Remote Write receiver and rule manager: Fix error handling. #11727
## 2.40.7 / 2022-12-14
* [BUGFIX] Use Windows native DNS resolver. #11704
* [BUGFIX] TSDB: Fix queries involving negative buckets of native histograms. #11699
## 2.40.6 / 2022-12-09
* [SECURITY] Security upgrade from go and upstream dependencies that include
security fixes to the net/http and os packages. #11691
## 2.40.5 / 2022-12-01
* [BUGFIX] TSDB: Fix queries involving native histograms due to improper reset of iterators. #11643
## 2.40.4 / 2022-11-29
* [SECURITY] Fix basic authentication bypass vulnerability (CVE-2022-46146). GHSA-4v48-4q5m-8vx4
## 2.40.3 / 2022-11-23
* [BUGFIX] TSDB: Fix compaction after a deletion is called. #11623
## 2.40.2 / 2022-11-16
* [BUGFIX] UI: Fix black-on-black metric name color in dark mode. #11572
## 2.40.1 / 2022-11-09
* [BUGFIX] TSDB: Fix alignment for atomic int64 for 32 bit architecture. #11547
* [BUGFIX] Scrape: Fix accept headers. #11552
## 2.40.0 / 2022-11-08
This release introduces an experimental, native way of representing and storing histograms.
It can be enabled in Prometheus via `--enable-feature=native-histograms` to accept native histograms.
Enabling native histograms will also switch the preferred exposition format to protobuf.
To instrument your application with native histograms, use the `main` branch of `client_golang` (this will change for the final release when v1.14.0 of client_golang will be out), and set the `NativeHistogramBucketFactor` in your `HistogramOpts` (`1.1` is a good starting point).
Your existing histograms won't switch to native histograms until `NativeHistogramBucketFactor` is set.
* [FEATURE] Add **experimental** support for native histograms. Enable with the flag `--enable-feature=native-histograms`. #11447
* [FEATURE] SD: Add service discovery for OVHcloud. #10802
* [ENHANCEMENT] Kubernetes SD: Use protobuf encoding. #11353
* [ENHANCEMENT] TSDB: Use golang.org/x/exp/slices for improved sorting speed. #11054 #11318 #11380
* [ENHANCEMENT] Consul SD: Add enterprise admin partitions. Adds `__meta_consul_partition` label. Adds `partition` config in `consul_sd_config`. #11482
* [BUGFIX] API: Fix API error codes for `/api/v1/labels` and `/api/v1/series`. #11356
## 2.39.2 / 2022-11-09
* [BUGFIX] TSDB: Fix alignment for atomic int64 for 32 bit architecture. #11547
## 2.39.1 / 2022-10-07
* [BUGFIX] Rules: Fix notifier relabel changing the labels on active alerts. #11427
## 2.39.0 / 2022-10-05
* [FEATURE] **experimental** TSDB: Add support for ingesting out-of-order samples. This is configured via `out_of_order_time_window` field in the config file; check config file docs for more info. #11075

21
CONTRIBUTING.md
View file

@ -64,10 +64,10 @@ To add or update a new dependency, use the `go get` command:
```bash
# Pick the latest tagged release.
go install example.com/some/module/pkg@latest
go get example.com/some/module/pkg@latest
# Pick a specific version.
go install example.com/some/module/pkg@vX.Y.Z
go get example.com/some/module/pkg@vX.Y.Z
```
Tidy up the `go.mod` and `go.sum` files:
@ -78,3 +78,20 @@ GO111MODULE=on go mod tidy
```
You have to commit the changes to `go.mod` and `go.sum` before submitting the pull request.
## Working with the PromQL parser
The PromQL parser grammar is located in `promql/parser/generated_parser.y` and it can be built using `make parser`.
The parser is built using [goyacc](https://pkg.go.dev/golang.org/x/tools/cmd/goyacc)
If doing some sort of debugging, then it is possible to add some verbose output. After generating the parser, then you
can modify the `./promql/parser/generated_parser.y.go` manually.
```golang
// As of writing this was somewhere around line 600.
var (
yyDebug = 0 // This can be be a number 0 -> 5.
yyErrorVerbose = false // This can be set to true.
)
```

2
MAINTAINERS.md
View file

@ -10,7 +10,7 @@ Julien Pivotto (<roidelapluie@prometheus.io> / @roidelapluie) and Levi Harrison
* `prometheus-mixin`: Björn Rabenstein (<beorn@grafana.com> / @beorn7)
* `storage`
* `remote`: Chris Marchbanks (<csmarchbanks@gmail.com> / @csmarchbanks), Callum Styan (<callumstyan@gmail.com> / @cstyan), Bartłomiej Płotka (<bwplotka@gmail.com> / @bwplotka), Tom Wilkie (<tom.wilkie@gmail.com> / @tomwilkie)
* `tsdb`: Ganesh Vernekar (<ganesh@grafana.com> / @codesome), Bartłomiej Płotka (<bwplotka@gmail.com> / @bwplotka)
* `tsdb`: Ganesh Vernekar (<ganesh@grafana.com> / @codesome), Bartłomiej Płotka (<bwplotka@gmail.com> / @bwplotka), Jesús Vázquez (<jesus.vazquez@grafana.com> / @jesusvazquez)
* `agent`: Robert Fratto (<robert.fratto@grafana.com> / @rfratto)
* `web`
* `ui`: Julius Volz (<julius.volz@gmail.com> / @juliusv)

24
Makefile
View file

@ -78,6 +78,17 @@ assets-tarball: assets
@echo '>> packaging assets'
scripts/package_assets.sh
# We only want to generate the parser when there's changes to the grammar.
.PHONY: parser
parser:
@echo ">> running goyacc to generate the .go file."
ifeq (, $(shell command -v goyacc > /dev/null))
@echo "goyacc not installed so skipping"
@echo "To install: go install golang.org/x/tools/cmd/goyacc@v0.6.0"
else
goyacc -o promql/parser/generated_parser.y.go promql/parser/generated_parser.y
endif
.PHONY: test
# If we only want to only test go code we have to change the test target
# which is called by all.
@ -90,8 +101,10 @@ endif
.PHONY: npm_licenses
npm_licenses: ui-install
@echo ">> bundling npm licenses"
rm -f $(REACT_APP_NPM_LICENSES_TARBALL)
find $(UI_NODE_MODULES_PATH) -iname "license*" | tar cfj $(REACT_APP_NPM_LICENSES_TARBALL) --transform 's/^/npm_licenses\//' --files-from=-
rm -f $(REACT_APP_NPM_LICENSES_TARBALL) npm_licenses
ln -s . npm_licenses
find npm_licenses/$(UI_NODE_MODULES_PATH) -iname "license*" | tar cfj $(REACT_APP_NPM_LICENSES_TARBALL) --files-from=-
rm -f npm_licenses
.PHONY: tarball
tarball: npm_licenses common-tarball
@ -107,7 +120,7 @@ plugins/plugins.go: plugins.yml plugins/generate.go
plugins: plugins/plugins.go
.PHONY: build
build: assets npm_licenses assets-compress common-build plugins
build: assets npm_licenses assets-compress plugins common-build
.PHONY: bench_tsdb
bench_tsdb: $(PROMU)
@ -120,3 +133,8 @@ bench_tsdb: $(PROMU)
@$(GO) tool pprof --alloc_space -svg $(PROMTOOL) $(TSDB_BENCHMARK_OUTPUT_DIR)/mem.prof > $(TSDB_BENCHMARK_OUTPUT_DIR)/memprof.alloc.svg
@$(GO) tool pprof -svg $(PROMTOOL) $(TSDB_BENCHMARK_OUTPUT_DIR)/block.prof > $(TSDB_BENCHMARK_OUTPUT_DIR)/blockprof.svg
@$(GO) tool pprof -svg $(PROMTOOL) $(TSDB_BENCHMARK_OUTPUT_DIR)/mutex.prof > $(TSDB_BENCHMARK_OUTPUT_DIR)/mutexprof.svg
.PHONY: cli-documentation
cli-documentation:
$(GO) run ./cmd/prometheus/ --write-documentation > docs/command-line/prometheus.md
$(GO) run ./cmd/promtool/ write-documentation > docs/command-line/promtool.md

22
Makefile.common
View file

@ -49,19 +49,19 @@ endif
GOTEST := $(GO) test
GOTEST_DIR :=
ifneq ($(CIRCLE_JOB),)
ifneq ($(shell which gotestsum),)
ifneq ($(shell command -v gotestsum > /dev/null),)
GOTEST_DIR := test-results
GOTEST := gotestsum --junitfile $(GOTEST_DIR)/unit-tests.xml --
endif
endif
PROMU_VERSION ?= 0.13.0
PROMU_VERSION ?= 0.15.0
PROMU_URL := https://github.com/prometheus/promu/releases/download/v$(PROMU_VERSION)/promu-$(PROMU_VERSION).$(GO_BUILD_PLATFORM).tar.gz
SKIP_GOLANGCI_LINT :=
GOLANGCI_LINT :=
GOLANGCI_LINT_OPTS ?=
GOLANGCI_LINT_VERSION ?= v1.49.0
GOLANGCI_LINT_VERSION ?= v1.53.3
# golangci-lint only supports linux, darwin and windows platforms on i386/amd64.
# windows isn't included here because of the path separator being different.
ifeq ($(GOHOSTOS),$(filter $(GOHOSTOS),linux darwin))
@ -91,6 +91,8 @@ BUILD_DOCKER_ARCHS = $(addprefix common-docker-,$(DOCKER_ARCHS))
PUBLISH_DOCKER_ARCHS = $(addprefix common-docker-publish-,$(DOCKER_ARCHS))
TAG_DOCKER_ARCHS = $(addprefix common-docker-tag-latest-,$(DOCKER_ARCHS))
SANITIZED_DOCKER_IMAGE_TAG := $(subst +,-,$(DOCKER_IMAGE_TAG))
ifeq ($(GOHOSTARCH),amd64)
ifeq ($(GOHOSTOS),$(filter $(GOHOSTOS),linux freebsd darwin windows))
# Only supported on amd64
@ -176,7 +178,7 @@ endif
.PHONY: common-yamllint
common-yamllint:
@echo ">> running yamllint on all YAML files in the repository"
ifeq (, $(shell which yamllint))
ifeq (, $(shell command -v yamllint > /dev/null))
@echo "yamllint not installed so skipping"
else
yamllint .
@ -205,7 +207,7 @@ common-tarball: promu
.PHONY: common-docker $(BUILD_DOCKER_ARCHS)
common-docker: $(BUILD_DOCKER_ARCHS)
$(BUILD_DOCKER_ARCHS): common-docker-%:
docker build -t "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$*:$(DOCKER_IMAGE_TAG)" \
docker build -t "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$*:$(SANITIZED_DOCKER_IMAGE_TAG)" \
-f $(DOCKERFILE_PATH) \
--build-arg ARCH="$*" \
--build-arg OS="linux" \
@ -214,19 +216,19 @@ $(BUILD_DOCKER_ARCHS): common-docker-%:
.PHONY: common-docker-publish $(PUBLISH_DOCKER_ARCHS)
common-docker-publish: $(PUBLISH_DOCKER_ARCHS)
$(PUBLISH_DOCKER_ARCHS): common-docker-publish-%:
docker push "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$*:$(DOCKER_IMAGE_TAG)"
docker push "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$*:$(SANITIZED_DOCKER_IMAGE_TAG)"
DOCKER_MAJOR_VERSION_TAG = $(firstword $(subst ., ,$(shell cat VERSION)))
.PHONY: common-docker-tag-latest $(TAG_DOCKER_ARCHS)
common-docker-tag-latest: $(TAG_DOCKER_ARCHS)
$(TAG_DOCKER_ARCHS): common-docker-tag-latest-%:
docker tag "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$*:$(DOCKER_IMAGE_TAG)" "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$*:latest"
docker tag "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$*:$(DOCKER_IMAGE_TAG)" "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$*:v$(DOCKER_MAJOR_VERSION_TAG)"
docker tag "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$*:$(SANITIZED_DOCKER_IMAGE_TAG)" "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$*:latest"
docker tag "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$*:$(SANITIZED_DOCKER_IMAGE_TAG)" "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$*:v$(DOCKER_MAJOR_VERSION_TAG)"
.PHONY: common-docker-manifest
common-docker-manifest:
DOCKER_CLI_EXPERIMENTAL=enabled docker manifest create -a "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME):$(DOCKER_IMAGE_TAG)" $(foreach ARCH,$(DOCKER_ARCHS),$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$(ARCH):$(DOCKER_IMAGE_TAG))
DOCKER_CLI_EXPERIMENTAL=enabled docker manifest push "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME):$(DOCKER_IMAGE_TAG)"
DOCKER_CLI_EXPERIMENTAL=enabled docker manifest create -a "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME):$(SANITIZED_DOCKER_IMAGE_TAG)" $(foreach ARCH,$(DOCKER_ARCHS),$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME)-linux-$(ARCH):$(SANITIZED_DOCKER_IMAGE_TAG))
DOCKER_CLI_EXPERIMENTAL=enabled docker manifest push "$(DOCKER_REPO)/$(DOCKER_IMAGE_NAME):$(SANITIZED_DOCKER_IMAGE_TAG)"
.PHONY: promu
promu: $(PROMU)

20
README.md
View file

@ -1,6 +1,13 @@
# Prometheus
<h1 align="center" style="border-bottom: none">
<a href="//prometheus.io" target="_blank"><img alt="Prometheus" src="/documentation/images/prometheus-logo.svg"></a><br>Prometheus
</h1>
[![CircleCI](https://circleci.com/gh/prometheus/prometheus/tree/main.svg?style=shield)][circleci]
<p align="center">Visit <a href="//prometheus.io" target="_blank">prometheus.io</a> for the full documentation,
examples and guides.</p>
<div align="center">
[![CI](https://github.com/prometheus/prometheus/actions/workflows/ci.yml/badge.svg)](https://github.com/prometheus/prometheus/actions/workflows/ci.yml)
[![Docker Repository on Quay](https://quay.io/repository/prometheus/prometheus/status)][quay]
[![Docker Pulls](https://img.shields.io/docker/pulls/prom/prometheus.svg?maxAge=604800)][hub]
[![Go Report Card](https://goreportcard.com/badge/github.com/prometheus/prometheus)](https://goreportcard.com/report/github.com/prometheus/prometheus)
@ -8,8 +15,7 @@
[![Gitpod ready-to-code](https://img.shields.io/badge/Gitpod-ready--to--code-blue?logo=gitpod)](https://gitpod.io/#https://github.com/prometheus/prometheus)
[![Fuzzing Status](https://oss-fuzz-build-logs.storage.googleapis.com/badges/prometheus.svg)](https://bugs.chromium.org/p/oss-fuzz/issues/list?sort=-opened&can=1&q=proj:prometheus)
Visit [prometheus.io](https://prometheus.io) for the full documentation,
examples and guides.
</div>
Prometheus, a [Cloud Native Computing Foundation](https://cncf.io/) project, is a systems and service monitoring system. It collects metrics
from configured targets at given intervals, evaluates rule expressions,
@ -28,7 +34,7 @@ The features that distinguish Prometheus from other metrics and monitoring syste
## Architecture overview
![Architecture overview](https://cdn.jsdelivr.net/gh/prometheus/prometheus@c34257d069c630685da35bcef084632ffd5d6209/documentation/images/architecture.svg)
![Architecture overview](documentation/images/architecture.svg)
## Install
@ -132,8 +138,6 @@ make npm_licenses
make common-docker-amd64
```
*NB* if you are on a Mac, you will need [gnu-tar](https://formulae.brew.sh/formula/gnu-tar).
## Using Prometheus as a Go Library
### Remote Write
@ -172,7 +176,6 @@ For more information on building, running, and developing on the React-based UI,
## More information
* Godoc documentation is available via [pkg.go.dev](https://pkg.go.dev/github.com/prometheus/prometheus). Due to peculiarities of Go Modules, v2.x.y will be displayed as v0.x.y.
* You will find a CircleCI configuration in [`.circleci/config.yml`](.circleci/config.yml).
* See the [Community page](https://prometheus.io/community) for how to reach the Prometheus developers and users on various communication channels.
## Contributing
@ -184,5 +187,4 @@ Refer to [CONTRIBUTING.md](https://github.com/prometheus/prometheus/blob/main/CO
Apache License 2.0, see [LICENSE](https://github.com/prometheus/prometheus/blob/main/LICENSE).
[hub]: https://hub.docker.com/r/prom/prometheus/
[circleci]: https://circleci.com/gh/prometheus/prometheus
[quay]: https://quay.io/repository/prometheus/prometheus

11
RELEASE.md
View file

@ -44,8 +44,15 @@ Release cadence of first pre-releases being cut is 6 weeks.
| v2.37 LTS | 2022-06-29 | Julien Pivotto (GitHub: @roidelapluie) |
| v2.38 | 2022-08-10 | Julius Volz (GitHub: @juliusv) |
| v2.39 | 2022-09-21 | Ganesh Vernekar (GitHub: @codesome) |
| v2.40 | 2022-11-02 | **searching for volunteer** |
| v2.41 | 2022-12-14 | **searching for volunteer** |
| v2.40 | 2022-11-02 | Ganesh Vernekar (GitHub: @codesome) |
| v2.41 | 2022-12-14 | Julien Pivotto (GitHub: @roidelapluie) |
| v2.42 | 2023-01-25 | Kemal Akkoyun (GitHub: @kakkoyun) |
| v2.43 | 2023-03-08 | Julien Pivotto (GitHub: @roidelapluie) |
| v2.44 | 2023-04-19 | Bryan Boreham (GitHub: @bboreham) |
| v2.45 LTS | 2023-05-31 | Jesus Vazquez (Github: @jesusvazquez) |
| v2.46 | 2023-07-12 | Julien Pivotto (GitHub: @roidelapluie) |
| v2.47 | 2023-08-23 | **searching for volunteer** |
| v2.48 | 2023-10-04 | **searching for volunteer** |
If you are interested in volunteering please create a pull request against the [prometheus/prometheus](https://github.com/prometheus/prometheus) repository and propose yourself for the release series of your choice.

2
VERSION
View file

@ -1 +1 @@
2.39.0
2.45.0

112
cmd/prometheus/main.go
View file

@ -12,6 +12,7 @@
// limitations under the License.
// The main package for the Prometheus server executable.
// nolint:revive // Many unsued function arguments in this file by design.
package main
import (
@ -33,6 +34,7 @@ import (
"syscall"
"time"
"github.com/alecthomas/kingpin/v2"
"github.com/alecthomas/units"
"github.com/go-kit/log"
"github.com/go-kit/log/level"
@ -45,10 +47,8 @@ import (
promlogflag "github.com/prometheus/common/promlog/flag"
"github.com/prometheus/common/version"
toolkit_web "github.com/prometheus/exporter-toolkit/web"
toolkit_webflag "github.com/prometheus/exporter-toolkit/web/kingpinflag"
"go.uber.org/atomic"
"go.uber.org/automaxprocs/maxprocs"
"gopkg.in/alecthomas/kingpin.v2"
"k8s.io/klog"
klogv2 "k8s.io/klog/v2"
@ -57,6 +57,7 @@ import (
"github.com/prometheus/prometheus/discovery/legacymanager"
"github.com/prometheus/prometheus/discovery/targetgroup"
"github.com/prometheus/prometheus/model/exemplar"
"github.com/prometheus/prometheus/model/histogram"
"github.com/prometheus/prometheus/model/labels"
"github.com/prometheus/prometheus/model/metadata"
"github.com/prometheus/prometheus/model/relabel"
@ -70,9 +71,10 @@ import (
"github.com/prometheus/prometheus/tracing"
"github.com/prometheus/prometheus/tsdb"
"github.com/prometheus/prometheus/tsdb/agent"
"github.com/prometheus/prometheus/tsdb/wlog"
"github.com/prometheus/prometheus/util/documentcli"
"github.com/prometheus/prometheus/util/logging"
prom_runtime "github.com/prometheus/prometheus/util/runtime"
"github.com/prometheus/prometheus/util/strutil"
"github.com/prometheus/prometheus/web"
)
@ -195,6 +197,10 @@ func (c *flagConfig) setFeatureListOptions(logger log.Logger) error {
case "no-default-scrape-port":
c.scrape.NoDefaultPort = true
level.Info(logger).Log("msg", "No default port will be appended to scrape targets' addresses.")
case "native-histograms":
c.tsdb.EnableNativeHistograms = true
c.scrape.EnableProtobufNegotiation = true
level.Info(logger).Log("msg", "Experimental native histogram support enabled.")
case "":
continue
case "promql-at-modifier", "promql-negative-offset":
@ -204,6 +210,12 @@ func (c *flagConfig) setFeatureListOptions(logger log.Logger) error {
}
}
}
if c.tsdb.EnableNativeHistograms && c.tsdb.EnableMemorySnapshotOnShutdown {
c.tsdb.EnableMemorySnapshotOnShutdown = false
level.Warn(logger).Log("msg", "memory-snapshot-on-shutdown has been disabled automatically because memory-snapshot-on-shutdown and native-histograms cannot be enabled at the same time.")
}
return nil
}
@ -241,7 +253,10 @@ func main() {
a.Flag("web.listen-address", "Address to listen on for UI, API, and telemetry.").
Default("0.0.0.0:9090").StringVar(&cfg.web.ListenAddress)
webConfig := toolkit_webflag.AddFlags(a)
webConfig := a.Flag(
"web.config.file",
"[EXPERIMENTAL] Path to configuration file that can enable TLS or authentication.",
).Default("").String()
a.Flag("web.read-timeout",
"Maximum duration before timing out read of the request, and closing idle connections.").
@ -320,9 +335,15 @@ func main() {
serverOnlyFlag(a, "storage.tsdb.wal-compression", "Compress the tsdb WAL.").
Hidden().Default("true").BoolVar(&cfg.tsdb.WALCompression)
serverOnlyFlag(a, "storage.tsdb.wal-compression-type", "Compression algorithm for the tsdb WAL.").
Hidden().Default(string(wlog.CompressionSnappy)).EnumVar(&cfg.tsdb.WALCompressionType, string(wlog.CompressionSnappy), string(wlog.CompressionZstd))
serverOnlyFlag(a, "storage.tsdb.head-chunks-write-queue-size", "Size of the queue through which head chunks are written to the disk to be m-mapped, 0 disables the queue completely. Experimental.").
Default("0").IntVar(&cfg.tsdb.HeadChunksWriteQueueSize)
serverOnlyFlag(a, "storage.tsdb.samples-per-chunk", "Target number of samples per chunk.").
Default("120").Hidden().IntVar(&cfg.tsdb.SamplesPerChunk)
agentOnlyFlag(a, "storage.agent.path", "Base path for metrics storage.").
Default("data-agent/").StringVar(&cfg.agentStoragePath)
@ -333,6 +354,9 @@ func main() {
agentOnlyFlag(a, "storage.agent.wal-compression", "Compress the agent WAL.").
Default("true").BoolVar(&cfg.agent.WALCompression)
agentOnlyFlag(a, "storage.agent.wal-compression-type", "Compression algorithm for the agent WAL.").
Hidden().Default(string(wlog.CompressionSnappy)).EnumVar(&cfg.agent.WALCompressionType, string(wlog.CompressionSnappy), string(wlog.CompressionZstd))
agentOnlyFlag(a, "storage.agent.wal-truncate-frequency",
"The frequency at which to truncate the WAL and remove old data.").
Hidden().PlaceHolder("<duration>").SetValue(&cfg.agent.TruncateFrequency)
@ -396,14 +420,23 @@ func main() {
a.Flag("scrape.discovery-reload-interval", "Interval used by scrape manager to throttle target groups updates.").
Hidden().Default("5s").SetValue(&cfg.scrape.DiscoveryReloadInterval)
a.Flag("enable-feature", "Comma separated feature names to enable. Valid options: agent, exemplar-storage, expand-external-labels, memory-snapshot-on-shutdown, promql-at-modifier, promql-negative-offset, promql-per-step-stats, remote-write-receiver (DEPRECATED), extra-scrape-metrics, new-service-discovery-manager, auto-gomaxprocs, no-default-scrape-port. See https://prometheus.io/docs/prometheus/latest/feature_flags/ for more details.").
a.Flag("enable-feature", "Comma separated feature names to enable. Valid options: agent, exemplar-storage, expand-external-labels, memory-snapshot-on-shutdown, promql-at-modifier, promql-negative-offset, promql-per-step-stats, remote-write-receiver (DEPRECATED), extra-scrape-metrics, new-service-discovery-manager, auto-gomaxprocs, no-default-scrape-port, native-histograms. See https://prometheus.io/docs/prometheus/latest/feature_flags/ for more details.").
Default("").StringsVar(&cfg.featureList)
promlogflag.AddFlags(a, &cfg.promlogConfig)
a.Flag("write-documentation", "Generate command line documentation. Internal use.").Hidden().Action(func(ctx *kingpin.ParseContext) error {
if err := documentcli.GenerateMarkdown(a.Model(), os.Stdout); err != nil {
os.Exit(1)
return err
}
os.Exit(0)
return nil
}).Bool()
_, err := a.Parse(os.Args[1:])
if err != nil {
fmt.Fprintln(os.Stderr, fmt.Errorf("Error parsing commandline arguments: %w", err))
fmt.Fprintln(os.Stderr, fmt.Errorf("Error parsing command line arguments: %w", err))
a.Usage(os.Args[1:])
os.Exit(2)
}
@ -456,11 +489,19 @@ func main() {
level.Error(logger).Log("msg", fmt.Sprintf("Error loading config (--config.file=%s)", cfg.configFile), "file", absPath, "err", err)
os.Exit(2)
}
if _, err := cfgFile.GetScrapeConfigs(); err != nil {
absPath, pathErr := filepath.Abs(cfg.configFile)
if pathErr != nil {
absPath = cfg.configFile
}
level.Error(logger).Log("msg", fmt.Sprintf("Error loading scrape config files from config (--config.file=%q)", cfg.configFile), "file", absPath, "err", err)
os.Exit(2)
}
if cfg.tsdb.EnableExemplarStorage {
if cfgFile.StorageConfig.ExemplarsConfig == nil {
cfgFile.StorageConfig.ExemplarsConfig = &config.DefaultExemplarsConfig
}
cfg.tsdb.MaxExemplars = int64(cfgFile.StorageConfig.ExemplarsConfig.MaxExemplars)
cfg.tsdb.MaxExemplars = cfgFile.StorageConfig.ExemplarsConfig.MaxExemplars
}
if cfgFile.StorageConfig.TSDBConfig != nil {
cfg.tsdb.OutOfOrderTimeWindow = cfgFile.StorageConfig.TSDBConfig.OutOfOrderTimeWindow
@ -619,7 +660,7 @@ func main() {
Appendable: fanoutStorage,
Queryable: localStorage,
QueryFunc: rules.EngineQueryFunc(queryEngine, fanoutStorage),
NotifyFunc: sendAlerts(notifierManager, cfg.web.ExternalURL.String()),
NotifyFunc: rules.SendAlerts(notifierManager, cfg.web.ExternalURL.String()),
Context: ctxRule,
ExternalURL: cfg.web.ExternalURL,
Registerer: prometheus.DefaultRegisterer,
@ -718,7 +759,11 @@ func main() {
name: "scrape_sd",
reloader: func(cfg *config.Config) error {
c := make(map[string]discovery.Configs)
for _, v := range cfg.ScrapeConfigs {
scfgs, err := cfg.GetScrapeConfigs()
if err != nil {
return err
}
for _, v := range scfgs {
c[v.JobName] = v.ServiceDiscoveryConfigs
}
return discoveryManagerScrape.ApplyConfig(c)
@ -1015,6 +1060,7 @@ func main() {
startTimeMargin := int64(2 * time.Duration(cfg.tsdb.MinBlockDuration).Seconds() * 1000)
localStorage.Set(db, startTimeMargin)
db.SetWriteNotified(remoteStorage)
close(dbOpen)
<-cancel
return nil
@ -1270,36 +1316,6 @@ func computeExternalURL(u, listenAddr string) (*url.URL, error) {
return eu, nil
}
type sender interface {
Send(alerts ...*notifier.Alert)
}
// sendAlerts implements the rules.NotifyFunc for a Notifier.
func sendAlerts(s sender, externalURL string) rules.NotifyFunc {
return func(ctx context.Context, expr string, alerts ...*rules.Alert) {
var res []*notifier.Alert
for _, alert := range alerts {
a := &notifier.Alert{
StartsAt: alert.FiredAt,
Labels: alert.Labels,
Annotations: alert.Annotations,
GeneratorURL: externalURL + strutil.TableLinkForExpression(expr),
}
if !alert.ResolvedAt.IsZero() {
a.EndsAt = alert.ResolvedAt
} else {
a.EndsAt = alert.ValidUntil
}
res = append(res, a)
}
if len(alerts) > 0 {
s.Send(res...)
}
}
}
// readyStorage implements the Storage interface while allowing to set the actual
// storage at a later point in time.
type readyStorage struct {
@ -1411,6 +1427,10 @@ func (n notReadyAppender) AppendExemplar(ref storage.SeriesRef, l labels.Labels,
return 0, tsdb.ErrNotReady
}
func (n notReadyAppender) AppendHistogram(ref storage.SeriesRef, l labels.Labels, t int64, h *histogram.Histogram, fh *histogram.FloatHistogram) (storage.SeriesRef, error) {
return 0, tsdb.ErrNotReady
}
func (n notReadyAppender) UpdateMetadata(ref storage.SeriesRef, l labels.Labels, m metadata.Metadata) (storage.SeriesRef, error) {
return 0, tsdb.ErrNotReady
}
@ -1473,11 +1493,11 @@ func (s *readyStorage) Snapshot(dir string, withHead bool) error {
}
// Stats implements the api_v1.TSDBAdminStats interface.
func (s *readyStorage) Stats(statsByLabelName string) (*tsdb.Stats, error) {
func (s *readyStorage) Stats(statsByLabelName string, limit int) (*tsdb.Stats, error) {
if x := s.get(); x != nil {
switch db := x.(type) {
case *tsdb.DB:
return db.Head().Stats(statsByLabelName), nil
return db.Head().Stats(statsByLabelName, limit), nil
case *agent.DB:
return nil, agent.ErrUnsupported
default:
@ -1533,7 +1553,9 @@ type tsdbOptions struct {
MaxBytes units.Base2Bytes
NoLockfile bool
WALCompression bool
WALCompressionType string
HeadChunksWriteQueueSize int
SamplesPerChunk int
StripeSize int
MinBlockDuration model.Duration
MaxBlockDuration model.Duration
@ -1541,6 +1563,7 @@ type tsdbOptions struct {
EnableExemplarStorage bool
MaxExemplars int64
EnableMemorySnapshotOnShutdown bool
EnableNativeHistograms bool
}
func (opts tsdbOptions) ToTSDBOptions() tsdb.Options {
@ -1551,14 +1574,16 @@ func (opts tsdbOptions) ToTSDBOptions() tsdb.Options {
MaxBytes: int64(opts.MaxBytes),
NoLockfile: opts.NoLockfile,
AllowOverlappingCompaction: true,
WALCompression: opts.WALCompression,
WALCompression: wlog.ParseCompressionType(opts.WALCompression, opts.WALCompressionType),
HeadChunksWriteQueueSize: opts.HeadChunksWriteQueueSize,
SamplesPerChunk: opts.SamplesPerChunk,
StripeSize: opts.StripeSize,
MinBlockDuration: int64(time.Duration(opts.MinBlockDuration) / time.Millisecond),
MaxBlockDuration: int64(time.Duration(opts.MaxBlockDuration) / time.Millisecond),
EnableExemplarStorage: opts.EnableExemplarStorage,
MaxExemplars: opts.MaxExemplars,
EnableMemorySnapshotOnShutdown: opts.EnableMemorySnapshotOnShutdown,
EnableNativeHistograms: opts.EnableNativeHistograms,
OutOfOrderTimeWindow: opts.OutOfOrderTimeWindow,
}
}
@ -1568,6 +1593,7 @@ func (opts tsdbOptions) ToTSDBOptions() tsdb.Options {
type agentOptions struct {
WALSegmentSize units.Base2Bytes
WALCompression bool
WALCompressionType string
StripeSize int
TruncateFrequency model.Duration
MinWALTime, MaxWALTime model.Duration
@ -1577,7 +1603,7 @@ type agentOptions struct {
func (opts agentOptions) ToAgentOptions() agent.Options {
return agent.Options{
WALSegmentSize: int(opts.WALSegmentSize),
WALCompression: opts.WALCompression,
WALCompression: wlog.ParseCompressionType(opts.WALCompression, opts.WALCompressionType),
StripeSize: opts.StripeSize,
TruncateFrequency: time.Duration(opts.TruncateFrequency),
MinWALTime: durationToInt64Millis(time.Duration(opts.MinWALTime)),

41
cmd/prometheus/main_test.go
View file

@ -23,6 +23,7 @@ import (
"os"
"os/exec"
"path/filepath"
"runtime"
"strings"
"syscall"
"testing"
@ -120,7 +121,7 @@ func TestFailedStartupExitCode(t *testing.T) {
fakeInputFile := "fake-input-file"
expectedExitStatus := 2
prom := exec.Command(promPath, "-test.main", "--config.file="+fakeInputFile)
prom := exec.Command(promPath, "-test.main", "--web.listen-address=0.0.0.0:0", "--config.file="+fakeInputFile)
err := prom.Run()
require.Error(t, err)
@ -198,7 +199,7 @@ func TestSendAlerts(t *testing.T) {
}
require.Equal(t, tc.exp, alerts)
})
sendAlerts(senderFunc, "http://localhost:9090")(context.TODO(), "up", tc.in...)
rules.SendAlerts(senderFunc, "http://localhost:9090")(context.TODO(), "up", tc.in...)
})
}
}
@ -357,7 +358,7 @@ func getCurrentGaugeValuesFor(t *testing.T, reg prometheus.Gatherer, metricNames
}
func TestAgentSuccessfulStartup(t *testing.T) {
prom := exec.Command(promPath, "-test.main", "--enable-feature=agent", "--config.file="+agentConfig)
prom := exec.Command(promPath, "-test.main", "--enable-feature=agent", "--web.listen-address=0.0.0.0:0", "--config.file="+agentConfig)
require.NoError(t, prom.Start())
actualExitStatus := 0
@ -375,7 +376,7 @@ func TestAgentSuccessfulStartup(t *testing.T) {
}
func TestAgentFailedStartupWithServerFlag(t *testing.T) {
prom := exec.Command(promPath, "-test.main", "--enable-feature=agent", "--storage.tsdb.path=.", "--config.file="+promConfig)
prom := exec.Command(promPath, "-test.main", "--enable-feature=agent", "--storage.tsdb.path=.", "--web.listen-address=0.0.0.0:0", "--config.file="+promConfig)
output := bytes.Buffer{}
prom.Stderr = &output
@ -402,7 +403,7 @@ func TestAgentFailedStartupWithServerFlag(t *testing.T) {
}
func TestAgentFailedStartupWithInvalidConfig(t *testing.T) {
prom := exec.Command(promPath, "-test.main", "--enable-feature=agent", "--config.file="+promConfig)
prom := exec.Command(promPath, "-test.main", "--enable-feature=agent", "--web.listen-address=0.0.0.0:0", "--config.file="+promConfig)
require.NoError(t, prom.Start())
actualExitStatus := 0
@ -437,7 +438,7 @@ func TestModeSpecificFlags(t *testing.T) {
for _, tc := range testcases {
t.Run(fmt.Sprintf("%s mode with option %s", tc.mode, tc.arg), func(t *testing.T) {
args := []string{"-test.main", tc.arg, t.TempDir()}
args := []string{"-test.main", tc.arg, t.TempDir(), "--web.listen-address=0.0.0.0:0"}
if tc.mode == "agent" {
args = append(args, "--enable-feature=agent", "--config.file="+agentConfig)
@ -483,3 +484,31 @@ func TestModeSpecificFlags(t *testing.T) {
})
}
}
func TestDocumentation(t *testing.T) {
if runtime.GOOS == "windows" {
t.SkipNow()
}
ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
defer cancel()
cmd := exec.CommandContext(ctx, promPath, "-test.main", "--write-documentation")
var stdout bytes.Buffer
cmd.Stdout = &stdout
if err := cmd.Run(); err != nil {
if exitError, ok := err.(*exec.ExitError); ok {
if exitError.ExitCode() != 0 {
fmt.Println("Command failed with non-zero exit code")
}
}
}
generatedContent := strings.ReplaceAll(stdout.String(), filepath.Base(promPath), strings.TrimSuffix(filepath.Base(promPath), ".test"))
expectedContent, err := os.ReadFile(filepath.Join("..", "..", "docs", "command-line", "prometheus.md"))
require.NoError(t, err)
require.Equal(t, string(expectedContent), generatedContent, "Generated content does not match documentation. Hint: run `make cli-documentation`.")
}

6
cmd/prometheus/main_unix_test.go
View file

@ -72,9 +72,11 @@ Loop:
if !startedOk {
t.Fatal("prometheus didn't start in the specified timeout")
}
if err := prom.Process.Kill(); err == nil {
switch err := prom.Process.Kill(); {
case err == nil:
t.Errorf("prometheus didn't shutdown gracefully after sending the Interrupt signal")
} else if stoppedErr != nil && stoppedErr.Error() != "signal: interrupt" { // TODO - find a better way to detect when the process didn't exit as expected!
case stoppedErr != nil && stoppedErr.Error() != "signal: interrupt":
// TODO: find a better way to detect when the process didn't exit as expected!
t.Errorf("prometheus exited with an unexpected error: %v", stoppedErr)
}
}

2
cmd/prometheus/query_log_test.go
View file

@ -193,7 +193,7 @@ func (p *queryLogTest) String() string {
}
name = name + ", " + p.host + ":" + strconv.Itoa(p.port)
if p.enabledAtStart {
name = name + ", enabled at start"
name += ", enabled at start"
}
if p.prefix != "" {
name = name + ", with prefix " + p.prefix

2
cmd/promtool/backfill.go
View file

@ -101,7 +101,7 @@ func createBlocks(input []byte, mint, maxt, maxBlockDuration int64, maxSamplesIn
nextSampleTs int64 = math.MaxInt64
)
for t := mint; t <= maxt; t = t + blockDuration {
for t := mint; t <= maxt; t += blockDuration {
tsUpper := t + blockDuration
if nextSampleTs != math.MaxInt64 && nextSampleTs >= tsUpper {
// The next sample is not in this timerange, we can avoid parsing

7
cmd/promtool/backfill_test.go
View file

@ -25,6 +25,7 @@ import (
"github.com/prometheus/prometheus/model/labels"
"github.com/prometheus/prometheus/storage"
"github.com/prometheus/prometheus/tsdb"
"github.com/prometheus/prometheus/tsdb/chunkenc"
)
type backfillSample struct {
@ -43,14 +44,14 @@ func sortSamples(samples []backfillSample) {
})
}
func queryAllSeries(t testing.TB, q storage.Querier, expectedMinTime, expectedMaxTime int64) []backfillSample {
func queryAllSeries(t testing.TB, q storage.Querier, expectedMinTime, expectedMaxTime int64) []backfillSample { // nolint:revive
ss := q.Select(false, nil, labels.MustNewMatcher(labels.MatchRegexp, "", ".*"))
samples := []backfillSample{}
for ss.Next() {
series := ss.At()
it := series.Iterator()
it := series.Iterator(nil)
require.NoError(t, it.Err())
for it.Next() {
for it.Next() == chunkenc.ValFloat {
ts, v := it.At()
samples = append(samples, backfillSample{Timestamp: ts, Value: v, Labels: series.Labels()})
}

264
cmd/promtool/main.go
View file

@ -31,6 +31,7 @@ import (
"text/tabwriter"
"time"
"github.com/alecthomas/kingpin/v2"
"github.com/go-kit/log"
"github.com/google/pprof/profile"
"github.com/prometheus/client_golang/api"
@ -41,10 +42,10 @@ import (
"github.com/prometheus/common/model"
"github.com/prometheus/common/version"
"github.com/prometheus/exporter-toolkit/web"
"gopkg.in/alecthomas/kingpin.v2"
"gopkg.in/yaml.v2"
dto "github.com/prometheus/client_model/go"
promconfig "github.com/prometheus/common/config"
"github.com/prometheus/common/expfmt"
"github.com/prometheus/prometheus/config"
@ -58,6 +59,7 @@ import (
_ "github.com/prometheus/prometheus/plugins" // Register plugins.
"github.com/prometheus/prometheus/promql"
"github.com/prometheus/prometheus/scrape"
"github.com/prometheus/prometheus/util/documentcli"
)
const (
@ -69,11 +71,20 @@ const (
lintOptionAll = "all"
lintOptionDuplicateRules = "duplicate-rules"
lintOptionNone = "none"
checkHealth = "/-/healthy"
checkReadiness = "/-/ready"
)
var lintOptions = []string{lintOptionAll, lintOptionDuplicateRules, lintOptionNone}
func main() {
var (
httpRoundTripper = api.DefaultRoundTripper
serverURL *url.URL
remoteWriteURL *url.URL
httpConfigFilePath string
)
app := kingpin.New(filepath.Base(os.Args[0]), "Tooling for the Prometheus monitoring system.").UsageWriter(os.Stdout)
app.Version(version.Print("promtool"))
app.HelpFlag.Short('h')
@ -105,11 +116,19 @@ func main() {
"The config files to check.",
).Required().ExistingFiles()
checkServerHealthCmd := checkCmd.Command("healthy", "Check if the Prometheus server is healthy.")
checkServerHealthCmd.Flag("http.config.file", "HTTP client configuration file for promtool to connect to Prometheus.").PlaceHolder("<filename>").ExistingFileVar(&httpConfigFilePath)
checkServerHealthCmd.Flag("url", "The URL for the Prometheus server.").Default("http://localhost:9090").URLVar(&serverURL)
checkServerReadyCmd := checkCmd.Command("ready", "Check if the Prometheus server is ready.")
checkServerReadyCmd.Flag("http.config.file", "HTTP client configuration file for promtool to connect to Prometheus.").PlaceHolder("<filename>").ExistingFileVar(&httpConfigFilePath)
checkServerReadyCmd.Flag("url", "The URL for the Prometheus server.").Default("http://localhost:9090").URLVar(&serverURL)
checkRulesCmd := checkCmd.Command("rules", "Check if the rule files are valid or not.")
ruleFiles := checkRulesCmd.Arg(
"rule-files",
"The rule files to check.",
).Required().ExistingFiles()
"The rule files to check, default is read from standard input.",
).ExistingFiles()
checkRulesLint := checkRulesCmd.Flag(
"lint",
"Linting checks to apply. Available options are: "+strings.Join(lintOptions, ", ")+". Use --lint=none to disable linting",
@ -124,14 +143,15 @@ func main() {
queryCmd := app.Command("query", "Run query against a Prometheus server.")
queryCmdFmt := queryCmd.Flag("format", "Output format of the query.").Short('o').Default("promql").Enum("promql", "json")
queryCmd.Flag("http.config.file", "HTTP client configuration file for promtool to connect to Prometheus.").PlaceHolder("<filename>").ExistingFileVar(&httpConfigFilePath)
queryInstantCmd := queryCmd.Command("instant", "Run instant query.")
queryInstantServer := queryInstantCmd.Arg("server", "Prometheus server to query.").Required().URL()
queryInstantCmd.Arg("server", "Prometheus server to query.").Required().URLVar(&serverURL)
queryInstantExpr := queryInstantCmd.Arg("expr", "PromQL query expression.").Required().String()
queryInstantTime := queryInstantCmd.Flag("time", "Query evaluation time (RFC3339 or Unix timestamp).").String()
queryRangeCmd := queryCmd.Command("range", "Run range query.")
queryRangeServer := queryRangeCmd.Arg("server", "Prometheus server to query.").Required().URL()
queryRangeCmd.Arg("server", "Prometheus server to query.").Required().URLVar(&serverURL)
queryRangeExpr := queryRangeCmd.Arg("expr", "PromQL query expression.").Required().String()
queryRangeHeaders := queryRangeCmd.Flag("header", "Extra headers to send to server.").StringMap()
queryRangeBegin := queryRangeCmd.Flag("start", "Query range start time (RFC3339 or Unix timestamp).").String()
@ -139,7 +159,7 @@ func main() {
queryRangeStep := queryRangeCmd.Flag("step", "Query step size (duration).").Duration()
querySeriesCmd := queryCmd.Command("series", "Run series query.")
querySeriesServer := querySeriesCmd.Arg("server", "Prometheus server to query.").Required().URL()
querySeriesCmd.Arg("server", "Prometheus server to query.").Required().URLVar(&serverURL)
querySeriesMatch := querySeriesCmd.Flag("match", "Series selector. Can be specified multiple times.").Required().Strings()
querySeriesBegin := querySeriesCmd.Flag("start", "Start time (RFC3339 or Unix timestamp).").String()
querySeriesEnd := querySeriesCmd.Flag("end", "End time (RFC3339 or Unix timestamp).").String()
@ -153,12 +173,24 @@ func main() {
debugAllServer := debugAllCmd.Arg("server", "Prometheus server to get all debug information from.").Required().String()
queryLabelsCmd := queryCmd.Command("labels", "Run labels query.")
queryLabelsServer := queryLabelsCmd.Arg("server", "Prometheus server to query.").Required().URL()
queryLabelsCmd.Arg("server", "Prometheus server to query.").Required().URLVar(&serverURL)
queryLabelsName := queryLabelsCmd.Arg("name", "Label name to provide label values for.").Required().String()
queryLabelsBegin := queryLabelsCmd.Flag("start", "Start time (RFC3339 or Unix timestamp).").String()
queryLabelsEnd := queryLabelsCmd.Flag("end", "End time (RFC3339 or Unix timestamp).").String()
queryLabelsMatch := queryLabelsCmd.Flag("match", "Series selector. Can be specified multiple times.").Strings()
pushCmd := app.Command("push", "Push to a Prometheus server.")
pushCmd.Flag("http.config.file", "HTTP client configuration file for promtool to connect to Prometheus.").PlaceHolder("<filename>").ExistingFileVar(&httpConfigFilePath)
pushMetricsCmd := pushCmd.Command("metrics", "Push metrics to a prometheus remote write (for testing purpose only).")
pushMetricsCmd.Arg("remote-write-url", "Prometheus remote write url to push metrics.").Required().URLVar(&remoteWriteURL)
metricFiles := pushMetricsCmd.Arg(
"metric-files",
"The metric files to push, default is read from standard input.",
).ExistingFiles()
pushMetricsLabels := pushMetricsCmd.Flag("label", "Label to attach to metrics. Can be specified multiple times.").Default("job=promtool").StringMap()
pushMetricsTimeout := pushMetricsCmd.Flag("timeout", "The time to wait for pushing metrics.").Default("30s").Duration()
pushMetricsHeaders := pushMetricsCmd.Flag("header", "Prometheus remote write header.").StringMap()
testCmd := app.Command("test", "Unit testing.")
testRulesCmd := testCmd.Command("rules", "Unit tests for rules.")
testRulesFiles := testRulesCmd.Arg(
@ -190,6 +222,7 @@ func main() {
dumpPath := tsdbDumpCmd.Arg("db path", "Database path (default is "+defaultDBPath+").").Default(defaultDBPath).String()
dumpMinTime := tsdbDumpCmd.Flag("min-time", "Minimum timestamp to dump.").Default(strconv.FormatInt(math.MinInt64, 10)).Int64()
dumpMaxTime := tsdbDumpCmd.Flag("max-time", "Maximum timestamp to dump.").Default(strconv.FormatInt(math.MaxInt64, 10)).Int64()
dumpMatch := tsdbDumpCmd.Flag("match", "Series selector.").Default("{__name__=~'(?s:.*)'}").String()
importCmd := tsdbCmd.Command("create-blocks-from", "[Experimental] Import samples from input and produce TSDB blocks. Please refer to the storage docs for more details.")
importHumanReadable := importCmd.Flag("human-readable", "Print human readable values.").Short('r').Bool()
@ -199,7 +232,8 @@ func main() {
importFilePath := openMetricsImportCmd.Arg("input file", "OpenMetrics file to read samples from.").Required().String()
importDBPath := openMetricsImportCmd.Arg("output directory", "Output directory for generated blocks.").Default(defaultDBPath).String()
importRulesCmd := importCmd.Command("rules", "Create blocks of data for new recording rules.")
importRulesURL := importRulesCmd.Flag("url", "The URL for the Prometheus API with the data where the rule will be backfilled from.").Default("http://localhost:9090").URL()
importRulesCmd.Flag("http.config.file", "HTTP client configuration file for promtool to connect to Prometheus.").PlaceHolder("<filename>").ExistingFileVar(&httpConfigFilePath)
importRulesCmd.Flag("url", "The URL for the Prometheus API with the data where the rule will be backfilled from.").Default("http://localhost:9090").URLVar(&serverURL)
importRulesStart := importRulesCmd.Flag("start", "The time to start backfilling the new rule from. Must be a RFC3339 formatted date or Unix timestamp. Required.").
Required().String()
importRulesEnd := importRulesCmd.Flag("end", "If an end time is provided, all recording rules in the rule files provided will be backfilled to the end time. Default will backfill up to 3 hours ago. Must be a RFC3339 formatted date or Unix timestamp.").String()
@ -213,6 +247,8 @@ func main() {
featureList := app.Flag("enable-feature", "Comma separated feature names to enable (only PromQL related and no-default-scrape-port). See https://prometheus.io/docs/prometheus/latest/feature_flags/ for the options and more details.").Default("").Strings()
documentationCmd := app.Command("write-documentation", "Generate command line documentation. Internal use.").Hidden()
parsedCmd := kingpin.MustParse(app.Parse(os.Args[1:]))
var p printer
@ -223,6 +259,22 @@ func main() {
p = &promqlPrinter{}
}
if httpConfigFilePath != "" {
if serverURL != nil && serverURL.User.Username() != "" {
kingpin.Fatalf("Cannot set base auth in the server URL and use a http.config.file at the same time")
}
var err error
httpConfig, _, err := config_util.LoadHTTPConfigFile(httpConfigFilePath)
if err != nil {
kingpin.Fatalf("Failed to load HTTP config file: %v", err)
}
httpRoundTripper, err = promconfig.NewRoundTripperFromConfig(*httpConfig, "promtool", config_util.WithUserAgent("promtool/"+version.Version))
if err != nil {
kingpin.Fatalf("Failed to create a new HTTP round tripper: %v", err)
}
}
var noDefaultScrapePort bool
for _, f := range *featureList {
opts := strings.Split(f, ",")
@ -247,6 +299,12 @@ func main() {
case checkConfigCmd.FullCommand():
os.Exit(CheckConfig(*agentMode, *checkConfigSyntaxOnly, newLintConfig(*checkConfigLint, *checkConfigLintFatal), *configFiles...))
case checkServerHealthCmd.FullCommand():
os.Exit(checkErr(CheckServerStatus(serverURL, checkHealth, httpRoundTripper)))
case checkServerReadyCmd.FullCommand():
os.Exit(checkErr(CheckServerStatus(serverURL, checkReadiness, httpRoundTripper)))
case checkWebConfigCmd.FullCommand():
os.Exit(CheckWebConfig(*webConfigFiles...))
@ -256,14 +314,17 @@ func main() {
case checkMetricsCmd.FullCommand():
os.Exit(CheckMetrics(*checkMetricsExtended))
case pushMetricsCmd.FullCommand():
os.Exit(PushMetrics(remoteWriteURL, httpRoundTripper, *pushMetricsHeaders, *pushMetricsTimeout, *pushMetricsLabels, *metricFiles...))
case queryInstantCmd.FullCommand():
os.Exit(QueryInstant(*queryInstantServer, *queryInstantExpr, *queryInstantTime, p))
os.Exit(QueryInstant(serverURL, httpRoundTripper, *queryInstantExpr, *queryInstantTime, p))
case queryRangeCmd.FullCommand():
os.Exit(QueryRange(*queryRangeServer, *queryRangeHeaders, *queryRangeExpr, *queryRangeBegin, *queryRangeEnd, *queryRangeStep, p))
os.Exit(QueryRange(serverURL, httpRoundTripper, *queryRangeHeaders, *queryRangeExpr, *queryRangeBegin, *queryRangeEnd, *queryRangeStep, p))
case querySeriesCmd.FullCommand():
os.Exit(QuerySeries(*querySeriesServer, *querySeriesMatch, *querySeriesBegin, *querySeriesEnd, p))
os.Exit(QuerySeries(serverURL, httpRoundTripper, *querySeriesMatch, *querySeriesBegin, *querySeriesEnd, p))
case debugPprofCmd.FullCommand():
os.Exit(debugPprof(*debugPprofServer))
@ -275,7 +336,7 @@ func main() {
os.Exit(debugAll(*debugAllServer))
case queryLabelsCmd.FullCommand():
os.Exit(QueryLabels(*queryLabelsServer, *queryLabelsMatch, *queryLabelsName, *queryLabelsBegin, *queryLabelsEnd, p))
os.Exit(QueryLabels(serverURL, httpRoundTripper, *queryLabelsMatch, *queryLabelsName, *queryLabelsBegin, *queryLabelsEnd, p))
case testRulesCmd.FullCommand():
os.Exit(RulesUnitTest(
@ -296,13 +357,15 @@ func main() {
os.Exit(checkErr(listBlocks(*listPath, *listHumanReadable)))
case tsdbDumpCmd.FullCommand():
os.Exit(checkErr(dumpSamples(*dumpPath, *dumpMinTime, *dumpMaxTime)))
os.Exit(checkErr(dumpSamples(*dumpPath, *dumpMinTime, *dumpMaxTime, *dumpMatch)))
// TODO(aSquare14): Work on adding support for custom block size.
case openMetricsImportCmd.FullCommand():
os.Exit(backfillOpenMetrics(*importFilePath, *importDBPath, *importHumanReadable, *importQuiet, *maxBlockDuration))
case importRulesCmd.FullCommand():
os.Exit(checkErr(importRules(*importRulesURL, *importRulesStart, *importRulesEnd, *importRulesOutputDir, *importRulesEvalInterval, *maxBlockDuration, *importRulesFiles...)))
os.Exit(checkErr(importRules(serverURL, httpRoundTripper, *importRulesStart, *importRulesEnd, *importRulesOutputDir, *importRulesEvalInterval, *maxBlockDuration, *importRulesFiles...)))
case documentationCmd.FullCommand():
os.Exit(checkErr(documentcli.GenerateMarkdown(app.Model(), os.Stdout)))
}
}
@ -338,6 +401,43 @@ func (ls lintConfig) lintDuplicateRules() bool {
return ls.all || ls.duplicateRules
}
// Check server status - healthy & ready.
func CheckServerStatus(serverURL *url.URL, checkEndpoint string, roundTripper http.RoundTripper) error {
if serverURL.Scheme == "" {
serverURL.Scheme = "http"
}
config := api.Config{
Address: serverURL.String() + checkEndpoint,
RoundTripper: roundTripper,
}
// Create new client.
c, err := api.NewClient(config)
if err != nil {
fmt.Fprintln(os.Stderr, "error creating API client:", err)
return err
}
request, err := http.NewRequest("GET", config.Address, nil)
if err != nil {
return err
}
ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
defer cancel()
response, dataBytes, err := c.Do(ctx, request)
if err != nil {
return err
}
if response.StatusCode != http.StatusOK {
return fmt.Errorf("check failed: URL=%s, status=%d", serverURL, response.StatusCode)
}
fmt.Fprintln(os.Stderr, " SUCCESS: ", string(dataBytes))
return nil
}
// CheckConfig validates configuration files.
func CheckConfig(agentMode, checkSyntaxOnly bool, lintSettings lintConfig, files ...string) int {
failed := false
@ -357,20 +457,12 @@ func CheckConfig(agentMode, checkSyntaxOnly bool, lintSettings lintConfig, files
}
fmt.Println()
for _, rf := range ruleFiles {
if n, errs := checkRules(rf, lintSettings); len(errs) > 0 {
fmt.Fprintln(os.Stderr, " FAILED:")
for _, err := range errs {
fmt.Fprintln(os.Stderr, " ", err)
}
failed = true
for _, err := range errs {
hasErrors = hasErrors || !errors.Is(err, lintError)
}
} else {
fmt.Printf(" SUCCESS: %d rules found\n", n)
}
fmt.Println()
rulesFailed, rulesHasErrors := checkRules(ruleFiles, lintSettings)
if rulesFailed {
failed = rulesFailed
}
if rulesHasErrors {
hasErrors = rulesHasErrors
}
}
if failed && hasErrors {
@ -437,7 +529,18 @@ func checkConfig(agentMode bool, filename string, checkSyntaxOnly bool) ([]strin
}
}
for _, scfg := range cfg.ScrapeConfigs {
var scfgs []*config.ScrapeConfig
if checkSyntaxOnly {
scfgs = cfg.ScrapeConfigs
} else {
var err error
scfgs, err = cfg.GetScrapeConfigs()
if err != nil {
return nil, fmt.Errorf("error loading scrape configs: %w", err)
}
}
for _, scfg := range scfgs {
if !checkSyntaxOnly && scfg.HTTPClientConfig.Authorization != nil {
if err := checkFileExists(scfg.HTTPClientConfig.Authorization.CredentialsFile); err != nil {
return nil, fmt.Errorf("error checking authorization credentials or bearer token file %q: %w", scfg.HTTPClientConfig.Authorization.CredentialsFile, err)
@ -587,9 +690,57 @@ func checkSDFile(filename string) ([]*targetgroup.Group, error) {
func CheckRules(ls lintConfig, files ...string) int {
failed := false
hasErrors := false
if len(files) == 0 {
fmt.Println("Checking standard input")
data, err := io.ReadAll(os.Stdin)
if err != nil {
fmt.Fprintln(os.Stderr, " FAILED:", err)
return failureExitCode
}
rgs, errs := rulefmt.Parse(data)
for _, e := range errs {
fmt.Fprintln(os.Stderr, e.Error())
return failureExitCode
}
if n, errs := checkRuleGroups(rgs, ls); errs != nil {
fmt.Fprintln(os.Stderr, " FAILED:")
for _, e := range errs {
fmt.Fprintln(os.Stderr, e.Error())
}
failed = true
for _, err := range errs {
hasErrors = hasErrors || !errors.Is(err, lintError)
}
} else {
fmt.Printf(" SUCCESS: %d rules found\n", n)
}
fmt.Println()
} else {
failed, hasErrors = checkRules(files, ls)
}
if failed && hasErrors {
return failureExitCode
}
if failed && ls.fatal {
return lintErrExitCode
}
return successExitCode
}
// checkRules validates rule files.
func checkRules(files []string, ls lintConfig) (bool, bool) {
failed := false
hasErrors := false
for _, f := range files {
if n, errs := checkRules(f, ls); errs != nil {
fmt.Println("Checking", f)
rgs, errs := rulefmt.ParseFile(f)
if errs != nil {
failed = true
continue
}
if n, errs := checkRuleGroups(rgs, ls); errs != nil {
fmt.Fprintln(os.Stderr, " FAILED:")
for _, e := range errs {
fmt.Fprintln(os.Stderr, e.Error())
@ -603,23 +754,10 @@ func CheckRules(ls lintConfig, files ...string) int {
}
fmt.Println()
}
if failed && hasErrors {
return failureExitCode
}
if failed && ls.fatal {
return lintErrExitCode
}
return successExitCode
return failed, hasErrors
}
func checkRules(filename string, lintSettings lintConfig) (int, []error) {
fmt.Println("Checking", filename)
rgs, errs := rulefmt.ParseFile(filename)
if errs != nil {
return successExitCode, errs
}
func checkRuleGroups(rgs *rulefmt.RuleGroups, lintSettings lintConfig) (int, []error) {
numRules := 0
for _, rg := range rgs.Groups {
numRules += len(rg.Rules)
@ -631,9 +769,9 @@ func checkRules(filename string, lintSettings lintConfig) (int, []error) {
errMessage := fmt.Sprintf("%d duplicate rule(s) found.\n", len(dRules))
for _, n := range dRules {
errMessage += fmt.Sprintf("Metric: %s\nLabel(s):\n", n.metric)
for _, l := range n.label {
n.label.Range(func(l labels.Label) {
errMessage += fmt.Sprintf("\t%s: %s\n", l.Name, l.Value)
}
})
}
errMessage += "Might cause inconsistency while recording expressions"
return 0, []error{fmt.Errorf("%w %s", lintError, errMessage)}
@ -794,12 +932,13 @@ func checkMetricsExtended(r io.Reader) ([]metricStat, int, error) {
}
// QueryInstant performs an instant query against a Prometheus server.
func QueryInstant(url *url.URL, query, evalTime string, p printer) int {
func QueryInstant(url *url.URL, roundTripper http.RoundTripper, query, evalTime string, p printer) int {
if url.Scheme == "" {
url.Scheme = "http"
}
config := api.Config{
Address: url.String(),
Address: url.String(),
RoundTripper: roundTripper,
}
// Create new client.
@ -834,12 +973,13 @@ func QueryInstant(url *url.URL, query, evalTime string, p printer) int {
}
// QueryRange performs a range query against a Prometheus server.
func QueryRange(url *url.URL, headers map[string]string, query, start, end string, step time.Duration, p printer) int {
func QueryRange(url *url.URL, roundTripper http.RoundTripper, headers map[string]string, query, start, end string, step time.Duration, p printer) int {
if url.Scheme == "" {
url.Scheme = "http"
}
config := api.Config{
Address: url.String(),
Address: url.String(),
RoundTripper: roundTripper,
}
if len(headers) > 0 {
@ -847,7 +987,7 @@ func QueryRange(url *url.URL, headers map[string]string, query, start, end strin
for key, value := range headers {
req.Header.Add(key, value)
}
return http.DefaultTransport.RoundTrip(req)
return roundTripper.RoundTrip(req)
})
}
@ -907,12 +1047,13 @@ func QueryRange(url *url.URL, headers map[string]string, query, start, end strin
}
// QuerySeries queries for a series against a Prometheus server.
func QuerySeries(url *url.URL, matchers []string, start, end string, p printer) int {
func QuerySeries(url *url.URL, roundTripper http.RoundTripper, matchers []string, start, end string, p printer) int {
if url.Scheme == "" {
url.Scheme = "http"
}
config := api.Config{
Address: url.String(),
Address: url.String(),
RoundTripper: roundTripper,
}
// Create new client.
@ -943,12 +1084,13 @@ func QuerySeries(url *url.URL, matchers []string, start, end string, p printer)
}
// QueryLabels queries for label values against a Prometheus server.
func QueryLabels(url *url.URL, matchers []string, name, start, end string, p printer) int {
func QueryLabels(url *url.URL, roundTripper http.RoundTripper, matchers []string, name, start, end string, p printer) int {
if url.Scheme == "" {
url.Scheme = "http"
}
config := api.Config{
Address: url.String(),
Address: url.String(),
RoundTripper: roundTripper,
}
// Create new client.
@ -1153,7 +1295,7 @@ func (j *jsonPrinter) printLabelValues(v model.LabelValues) {
// importRules backfills recording rules from the files provided. The output are blocks of data
// at the outputDir location.
func importRules(url *url.URL, start, end, outputDir string, evalInterval, maxBlockDuration time.Duration, files ...string) error {
func importRules(url *url.URL, roundTripper http.RoundTripper, start, end, outputDir string, evalInterval, maxBlockDuration time.Duration, files ...string) error {
ctx := context.Background()
var stime, etime time.Time
var err error
@ -1183,7 +1325,8 @@ func importRules(url *url.URL, start, end, outputDir string, evalInterval, maxBl
maxBlockDuration: maxBlockDuration,
}
client, err := api.NewClient(api.Config{
Address: url.String(),
Address: url.String(),
RoundTripper: roundTripper,
})
if err != nil {
return fmt.Errorf("new api client error: %w", err)
@ -1219,8 +1362,11 @@ func checkTargetGroupsForAlertmanager(targetGroups []*targetgroup.Group, amcfg *
}
func checkTargetGroupsForScrapeConfig(targetGroups []*targetgroup.Group, scfg *config.ScrapeConfig) error {
var targets []*scrape.Target
lb := labels.NewBuilder(labels.EmptyLabels())
for _, tg := range targetGroups {
_, failures := scrape.TargetsFromGroup(tg, scfg, false)
var failures []error
targets, failures = scrape.TargetsFromGroup(tg, scfg, false, targets, lb)
if len(failures) > 0 {
first := failures[0]
return first

37
cmd/promtool/main_test.go
View file

@ -14,6 +14,8 @@
package main
import (
"bytes"
"context"
"errors"
"fmt"
"net/http"
@ -21,6 +23,7 @@ import (
"net/url"
"os"
"os/exec"
"path/filepath"
"runtime"
"strings"
"syscall"
@ -56,14 +59,14 @@ func TestQueryRange(t *testing.T) {
require.Equal(t, nil, err)
p := &promqlPrinter{}
exitCode := QueryRange(urlObject, map[string]string{}, "up", "0", "300", 0, p)
exitCode := QueryRange(urlObject, http.DefaultTransport, map[string]string{}, "up", "0", "300", 0, p)
require.Equal(t, "/api/v1/query_range", getRequest().URL.Path)
form := getRequest().Form
require.Equal(t, "up", form.Get("query"))
require.Equal(t, "1", form.Get("step"))
require.Equal(t, 0, exitCode)
exitCode = QueryRange(urlObject, map[string]string{}, "up", "0", "300", 10*time.Millisecond, p)
exitCode = QueryRange(urlObject, http.DefaultTransport, map[string]string{}, "up", "0", "300", 10*time.Millisecond, p)
require.Equal(t, "/api/v1/query_range", getRequest().URL.Path)
form = getRequest().Form
require.Equal(t, "up", form.Get("query"))
@ -79,7 +82,7 @@ func TestQueryInstant(t *testing.T) {
require.Equal(t, nil, err)
p := &promqlPrinter{}
exitCode := QueryInstant(urlObject, "up", "300", p)
exitCode := QueryInstant(urlObject, http.DefaultTransport, "up", "300", p)
require.Equal(t, "/api/v1/query", getRequest().URL.Path)
form := getRequest().Form
require.Equal(t, "up", form.Get("query"))
@ -433,3 +436,31 @@ func TestExitCodes(t *testing.T) {
})
}
}
func TestDocumentation(t *testing.T) {
if runtime.GOOS == "windows" {
t.SkipNow()
}
ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
defer cancel()
cmd := exec.CommandContext(ctx, promtoolPath, "-test.main", "write-documentation")
var stdout bytes.Buffer
cmd.Stdout = &stdout
if err := cmd.Run(); err != nil {
if exitError, ok := err.(*exec.ExitError); ok {
if exitError.ExitCode() != 0 {
fmt.Println("Command failed with non-zero exit code")
}
}
}
generatedContent := strings.ReplaceAll(stdout.String(), filepath.Base(promtoolPath), strings.TrimSuffix(filepath.Base(promtoolPath), ".test"))
expectedContent, err := os.ReadFile(filepath.Join("..", "..", "docs", "command-line", "promtool.md"))
require.NoError(t, err)
require.Equal(t, string(expectedContent), generatedContent, "Generated content does not match documentation. Hint: run `make cli-documentation`.")
}

138
cmd/promtool/metrics.go Normal file
View file

@ -0,0 +1,138 @@
// Copyright 2023 The Prometheus Authors
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package main
import (
"bytes"
"context"
"fmt"
"io"
"net/http"
"net/url"
"os"
"time"
"github.com/golang/snappy"
config_util "github.com/prometheus/common/config"
"github.com/prometheus/common/model"
"github.com/prometheus/prometheus/storage/remote"
"github.com/prometheus/prometheus/util/fmtutil"
)
// Push metrics to a prometheus remote write (for testing purpose only).
func PushMetrics(url *url.URL, roundTripper http.RoundTripper, headers map[string]string, timeout time.Duration, labels map[string]string, files ...string) int {
addressURL, err := url.Parse(url.String())
if err != nil {
fmt.Fprintln(os.Stderr, err)
return failureExitCode
}
// build remote write client
writeClient, err := remote.NewWriteClient("remote-write", &remote.ClientConfig{
URL: &config_util.URL{URL: addressURL},
Timeout: model.Duration(timeout),
})
if err != nil {
fmt.Fprintln(os.Stderr, err)
return failureExitCode
}
// set custom tls config from httpConfigFilePath
// set custom headers to every request
client, ok := writeClient.(*remote.Client)
if !ok {
fmt.Fprintln(os.Stderr, fmt.Errorf("unexpected type %T", writeClient))
return failureExitCode
}
client.Client.Transport = &setHeadersTransport{
RoundTripper: roundTripper,
headers: headers,
}
var data []byte
var failed bool
if len(files) == 0 {
data, err = io.ReadAll(os.Stdin)
if err != nil {
fmt.Fprintln(os.Stderr, " FAILED:", err)
return failureExitCode
}
fmt.Printf("Parsing standard input\n")
if parseAndPushMetrics(client, data, labels) {
fmt.Printf(" SUCCESS: metrics pushed to remote write.\n")
return successExitCode
}
return failureExitCode
}
for _, file := range files {
data, err = os.ReadFile(file)
if err != nil {
fmt.Fprintln(os.Stderr, " FAILED:", err)
failed = true
continue
}
fmt.Printf("Parsing metrics file %s\n", file)
if parseAndPushMetrics(client, data, labels) {
fmt.Printf(" SUCCESS: metrics file %s pushed to remote write.\n", file)
continue
}
failed = true
}
if failed {
return failureExitCode
}
return successExitCode
}
func parseAndPushMetrics(client *remote.Client, data []byte, labels map[string]string) bool {
metricsData, err := fmtutil.MetricTextToWriteRequest(bytes.NewReader(data), labels)
if err != nil {
fmt.Fprintln(os.Stderr, " FAILED:", err)
return false
}
raw, err := metricsData.Marshal()
if err != nil {
fmt.Fprintln(os.Stderr, " FAILED:", err)
return false
}
// Encode the request body into snappy encoding.
compressed := snappy.Encode(nil, raw)
err = client.Store(context.Background(), compressed)
if err != nil {
fmt.Fprintln(os.Stderr, " FAILED:", err)
return false
}
return true
}
type setHeadersTransport struct {
http.RoundTripper
headers map[string]string
}
func (s *setHeadersTransport) RoundTrip(req *http.Request) (*http.Response, error) {
for key, value := range s.headers {
req.Header.Set(key, value)
}
return s.RoundTripper.RoundTrip(req)
}

11
cmd/promtool/rules.go
View file

@ -68,7 +68,7 @@ func newRuleImporter(logger log.Logger, config ruleImporterConfig, apiClient que
}
// loadGroups parses groups from a list of recording rule files.
func (importer *ruleImporter) loadGroups(ctx context.Context, filenames []string) (errs []error) {
func (importer *ruleImporter) loadGroups(_ context.Context, filenames []string) (errs []error) {
groups, errs := importer.ruleManager.LoadGroups(importer.config.evalInterval, labels.Labels{}, "", nil, filenames...)
if errs != nil {
return errs
@ -100,7 +100,7 @@ func (importer *ruleImporter) importRule(ctx context.Context, ruleExpr, ruleName
startInMs := start.Unix() * int64(time.Second/time.Millisecond)
endInMs := end.Unix() * int64(time.Second/time.Millisecond)
for startOfBlock := blockDuration * (startInMs / blockDuration); startOfBlock <= endInMs; startOfBlock = startOfBlock + blockDuration {
for startOfBlock := blockDuration * (startInMs / blockDuration); startOfBlock <= endInMs; startOfBlock += blockDuration {
endOfBlock := startOfBlock + blockDuration - 1
currStart := max(startOfBlock/int64(time.Second/time.Millisecond), start.Unix())
@ -158,14 +158,15 @@ func (importer *ruleImporter) importRule(ctx context.Context, ruleExpr, ruleName
// Setting the rule labels after the output of the query,
// so they can override query output.
for _, l := range ruleLabels {
ruleLabels.Range(func(l labels.Label) {
lb.Set(l.Name, l.Value)
}
})
lb.Set(labels.MetricName, ruleName)
lbls := lb.Labels()
for _, value := range sample.Values {
if err := app.add(ctx, lb.Labels(nil), timestamp.FromTime(value.Timestamp.Time()), float64(value.Value)); err != nil {
if err := app.add(ctx, lbls, timestamp.FromTime(value.Timestamp.Time()), float64(value.Value)); err != nil {
return fmt.Errorf("add: %w", err)
}
}

19
cmd/promtool/rules_test.go
View file

@ -28,13 +28,14 @@ import (
"github.com/prometheus/prometheus/model/labels"
"github.com/prometheus/prometheus/tsdb"
"github.com/prometheus/prometheus/tsdb/chunkenc"
)
type mockQueryRangeAPI struct {
samples model.Matrix
}
func (mockAPI mockQueryRangeAPI) QueryRange(ctx context.Context, query string, r v1.Range, opts ...v1.Option) (model.Value, v1.Warnings, error) {
func (mockAPI mockQueryRangeAPI) QueryRange(_ context.Context, query string, r v1.Range, opts ...v1.Option) (model.Value, v1.Warnings, error) { // nolint:revive
return mockAPI.samples, v1.Warnings{}, nil
}
@ -99,7 +100,7 @@ func TestBackfillRuleIntegration(t *testing.T) {
require.Equal(t, 1, len(gRules))
require.Equal(t, "rule1", gRules[0].Name())
require.Equal(t, "ruleExpr", gRules[0].Query().String())
require.Equal(t, 1, len(gRules[0].Labels()))
require.Equal(t, 1, gRules[0].Labels().Len())
group2 := ruleImporter.groups[path2+";group2"]
require.NotNil(t, group2)
@ -108,7 +109,7 @@ func TestBackfillRuleIntegration(t *testing.T) {
require.Equal(t, 2, len(g2Rules))
require.Equal(t, "grp2_rule1", g2Rules[0].Name())
require.Equal(t, "grp2_rule1_expr", g2Rules[0].Query().String())
require.Equal(t, 0, len(g2Rules[0].Labels()))
require.Equal(t, 0, g2Rules[0].Labels().Len())
// Backfill all recording rules then check the blocks to confirm the correct data was created.
errs = ruleImporter.importAll(ctx)
@ -131,15 +132,15 @@ func TestBackfillRuleIntegration(t *testing.T) {
for selectedSeries.Next() {
seriesCount++
series := selectedSeries.At()
if len(series.Labels()) != 3 {
require.Equal(t, 2, len(series.Labels()))
if series.Labels().Len() != 3 {
require.Equal(t, 2, series.Labels().Len())
x := labels.FromStrings("__name__", "grp2_rule1", "name1", "val1")
require.Equal(t, x, series.Labels())
} else {
require.Equal(t, 3, len(series.Labels()))
require.Equal(t, 3, series.Labels().Len())
}
it := series.Iterator()
for it.Next() {
it := series.Iterator(nil)
for it.Next() == chunkenc.ValFloat {
samplesCount++
ts, v := it.At()
if v == testValue {
@ -160,7 +161,7 @@ func TestBackfillRuleIntegration(t *testing.T) {
}
}
func newTestRuleImporter(ctx context.Context, start time.Time, tmpDir string, testSamples model.Matrix, maxBlockDuration time.Duration) (*ruleImporter, error) {
func newTestRuleImporter(_ context.Context, start time.Time, tmpDir string, testSamples model.Matrix, maxBlockDuration time.Duration) (*ruleImporter, error) {
logger := log.NewNopLogger()
cfg := ruleImporterConfig{
outputDir: tmpDir,

18
cmd/promtool/sd.go
View file

@ -47,9 +47,15 @@ func CheckSD(sdConfigFiles, sdJobName string, sdTimeout time.Duration, noDefault
}
var scrapeConfig *config.ScrapeConfig
scfgs, err := cfg.GetScrapeConfigs()
if err != nil {
fmt.Fprintln(os.Stderr, "Cannot load scrape configs", err)
return failureExitCode
}
jobs := []string{}
jobMatched := false
for _, v := range cfg.ScrapeConfigs {
for _, v := range scfgs {
jobs = append(jobs, v.JobName)
if v.JobName == sdJobName {
jobMatched = true
@ -109,22 +115,22 @@ outerLoop:
func getSDCheckResult(targetGroups []*targetgroup.Group, scrapeConfig *config.ScrapeConfig, noDefaultScrapePort bool) []sdCheckResult {
sdCheckResults := []sdCheckResult{}
lb := labels.NewBuilder(labels.EmptyLabels())
for _, targetGroup := range targetGroups {
for _, target := range targetGroup.Targets {
labelSlice := make([]labels.Label, 0, len(target)+len(targetGroup.Labels))
lb.Reset(labels.EmptyLabels())
for name, value := range target {
labelSlice = append(labelSlice, labels.Label{Name: string(name), Value: string(value)})
lb.Set(string(name), string(value))
}
for name, value := range targetGroup.Labels {
if _, ok := target[name]; !ok {
labelSlice = append(labelSlice, labels.Label{Name: string(name), Value: string(value)})
lb.Set(string(name), string(value))
}
}
targetLabels := labels.New(labelSlice...)
res, orig, err := scrape.PopulateLabels(targetLabels, scrapeConfig, noDefaultScrapePort)
res, orig, err := scrape.PopulateLabels(lb, scrapeConfig, noDefaultScrapePort)
result := sdCheckResult{
DiscoveredLabels: orig,
Labels: res,

71
cmd/promtool/tsdb.go
View file

@ -23,24 +23,25 @@ import (
"path/filepath"
"runtime"
"runtime/pprof"
"sort"
"strconv"
"strings"
"sync"
"text/tabwriter"
"time"
"github.com/prometheus/prometheus/storage"
"github.com/prometheus/prometheus/tsdb/index"
"github.com/alecthomas/units"
"github.com/go-kit/log"
"golang.org/x/exp/slices"
"github.com/prometheus/prometheus/model/labels"
"github.com/prometheus/prometheus/promql/parser"
"github.com/prometheus/prometheus/storage"
"github.com/prometheus/prometheus/tsdb"
"github.com/prometheus/prometheus/tsdb/chunkenc"
"github.com/prometheus/prometheus/tsdb/chunks"
tsdb_errors "github.com/prometheus/prometheus/tsdb/errors"
"github.com/prometheus/prometheus/tsdb/fileutil"
"github.com/prometheus/prometheus/tsdb/index"
)
const timeDelta = 30000
@ -314,7 +315,7 @@ func readPrometheusLabels(r io.Reader, n int) ([]labels.Labels, error) {
i := 0
for scanner.Scan() && i < n {
m := make(labels.Labels, 0, 10)
m := make([]labels.Label, 0, 10)
r := strings.NewReplacer("\"", "", "{", "", "}", "")
s := r.Replace(scanner.Text())
@ -324,13 +325,12 @@ func readPrometheusLabels(r io.Reader, n int) ([]labels.Labels, error) {
split := strings.Split(labelChunk, ":")
m = append(m, labels.Label{Name: split[0], Value: split[1]})
}
// Order of the k/v labels matters, don't assume we'll always receive them already sorted.
sort.Sort(m)
h := m.Hash()
ml := labels.New(m...) // This sorts by name - order of the k/v labels matters, don't assume we'll always receive them already sorted.
h := ml.Hash()
if _, ok := hashes[h]; ok {
continue
}
mets = append(mets, m)
mets = append(mets, ml)
hashes[h] = struct{}{}
i++
}
@ -397,25 +397,20 @@ func openBlock(path, blockID string) (*tsdb.DBReadOnly, tsdb.BlockReader, error)
if err != nil {
return nil, nil, err
}
blocks, err := db.Blocks()
if blockID == "" {
blockID, err = db.LastBlockID()
if err != nil {
return nil, nil, err
}
}
b, err := db.Block(blockID)
if err != nil {
return nil, nil, err
}
var block tsdb.BlockReader
if blockID != "" {
for _, b := range blocks {
if b.Meta().ULID.String() == blockID {
block = b
break
}
}
} else if len(blocks) > 0 {
block = blocks[len(blocks)-1]
}
if block == nil {
return nil, nil, fmt.Errorf("block %s not found", blockID)
}
return db, block, nil
return db, b, nil
}
func analyzeBlock(path, blockID string, limit int, runExtended bool) error {
@ -451,7 +446,7 @@ func analyzeBlock(path, blockID string, limit int, runExtended bool) error {
postingInfos := []postingInfo{}
printInfo := func(postingInfos []postingInfo) {
sort.Slice(postingInfos, func(i, j int) bool { return postingInfos[i].metric > postingInfos[j].metric })
slices.SortFunc(postingInfos, func(a, b postingInfo) bool { return a.metric > b.metric })
for i, pc := range postingInfos {
if i >= limit {
@ -469,21 +464,21 @@ func analyzeBlock(path, blockID string, limit int, runExtended bool) error {
if err != nil {
return err
}
lbls := labels.Labels{}
chks := []chunks.Meta{}
builder := labels.ScratchBuilder{}
for p.Next() {
if err = ir.Series(p.At(), &lbls, &chks); err != nil {
if err = ir.Series(p.At(), &builder, &chks); err != nil {
return err
}
// Amount of the block time range not covered by this series.
uncovered := uint64(meta.MaxTime-meta.MinTime) - uint64(chks[len(chks)-1].MaxTime-chks[0].MinTime)
for _, lbl := range lbls {
builder.Labels().Range(func(lbl labels.Label) {
key := lbl.Name + "=" + lbl.Value
labelsUncovered[lbl.Name] += uncovered
labelpairsUncovered[key] += uncovered
labelpairsCount[key]++
entries++
}
})
}
if p.Err() != nil {
return p.Err()
@ -588,10 +583,10 @@ func analyzeCompaction(block tsdb.BlockReader, indexr tsdb.IndexReader) (err err
nBuckets := 10
histogram := make([]int, nBuckets)
totalChunks := 0
var builder labels.ScratchBuilder
for postingsr.Next() {
lbsl := labels.Labels{}
var chks []chunks.Meta
if err := indexr.Series(postingsr.At(), &lbsl, &chks); err != nil {
if err := indexr.Series(postingsr.At(), &builder, &chks); err != nil {
return err
}
@ -624,7 +619,7 @@ func analyzeCompaction(block tsdb.BlockReader, indexr tsdb.IndexReader) (err err
return nil
}
func dumpSamples(path string, mint, maxt int64) (err error) {
func dumpSamples(path string, mint, maxt int64, match string) (err error) {
db, err := tsdb.OpenDBReadOnly(path, nil)
if err != nil {
return err
@ -638,13 +633,17 @@ func dumpSamples(path string, mint, maxt int64) (err error) {
}
defer q.Close()
ss := q.Select(false, nil, labels.MustNewMatcher(labels.MatchRegexp, "", ".*"))
matchers, err := parser.ParseMetricSelector(match)
if err != nil {
return err
}
ss := q.Select(false, nil, matchers...)
for ss.Next() {
series := ss.At()
lbs := series.Labels()
it := series.Iterator()
for it.Next() {
it := series.Iterator(nil)
for it.Next() == chunkenc.ValFloat {
ts, val := it.At()
fmt.Printf("%s %g %d\n", lbs, val, ts)
}

13
cmd/promtool/unittest.go
View file

@ -130,7 +130,7 @@ func resolveAndGlobFilepaths(baseDir string, utf *unitTestFile) error {
if err != nil {
return err
}
if len(m) <= 0 {
if len(m) == 0 {
fmt.Fprintln(os.Stderr, " WARNING: no file match pattern", rf)
}
globbedFiles = append(globbedFiles, m...)
@ -284,8 +284,8 @@ func (tg *testGroup) test(evalInterval time.Duration, groupOrderMap map[string]i
for _, a := range ar.ActiveAlerts() {
if a.State == rules.StateFiring {
alerts = append(alerts, labelAndAnnotation{
Labels: append(labels.Labels{}, a.Labels...),
Annotations: append(labels.Labels{}, a.Annotations...),
Labels: a.Labels.Copy(),
Annotations: a.Annotations.Copy(),
})
}
}
@ -347,7 +347,7 @@ Outer:
for _, s := range got {
gotSamples = append(gotSamples, parsedSample{
Labels: s.Metric.Copy(),
Value: s.V,
Value: s.F,
})
}
@ -434,7 +434,7 @@ func (tg *testGroup) maxEvalTime() time.Duration {
}
func query(ctx context.Context, qs string, t time.Time, engine *promql.Engine, qu storage.Queryable) (promql.Vector, error) {
q, err := engine.NewInstantQuery(qu, nil, qs, t)
q, err := engine.NewInstantQuery(ctx, qu, nil, qs, t)
if err != nil {
return nil, err
}
@ -447,7 +447,8 @@ func query(ctx context.Context, qs string, t time.Time, engine *promql.Engine, q
return v, nil
case promql.Scalar:
return promql.Vector{promql.Sample{
Point: promql.Point(v),
T: v.T,
F: v.V,
Metric: labels.Labels{},
}}, nil
default:

262
config/config.go
View file

@ -34,6 +34,7 @@ import (
"github.com/prometheus/prometheus/discovery"
"github.com/prometheus/prometheus/model/labels"
"github.com/prometheus/prometheus/model/relabel"
"github.com/prometheus/prometheus/storage/remote/azuread"
)
var (
@ -80,7 +81,8 @@ func Load(s string, expandExternalLabels bool, logger log.Logger) (*Config, erro
return cfg, nil
}
for i, v := range cfg.GlobalConfig.ExternalLabels {
b := labels.ScratchBuilder{}
cfg.GlobalConfig.ExternalLabels.Range(func(v labels.Label) {
newV := os.Expand(v.Value, func(s string) string {
if s == "$" {
return "$"
@ -93,10 +95,10 @@ func Load(s string, expandExternalLabels bool, logger log.Logger) (*Config, erro
})
if newV != v.Value {
level.Debug(logger).Log("msg", "External label replaced", "label", v.Name, "input", v.Value, "output", newV)
v.Value = newV
cfg.GlobalConfig.ExternalLabels[i] = v
}
}
b.Add(v.Name, newV)
})
cfg.GlobalConfig.ExternalLabels = b.Labels()
return cfg, nil
}
@ -112,10 +114,6 @@ func LoadFile(filename string, agentMode, expandExternalLabels bool, logger log.
}
if agentMode {
if len(cfg.RemoteWriteConfigs) == 0 {
return nil, errors.New("at least one remote_write target must be specified in agent mode")
}
if len(cfg.AlertingConfig.AlertmanagerConfigs) > 0 || len(cfg.AlertingConfig.AlertRelabelConfigs) > 0 {
return nil, errors.New("field alerting is not allowed in agent mode")
}
@ -149,13 +147,14 @@ var (
// DefaultScrapeConfig is the default scrape configuration.
DefaultScrapeConfig = ScrapeConfig{
// ScrapeTimeout and ScrapeInterval default to the
// configured globals.
MetricsPath: "/metrics",
Scheme: "http",
HonorLabels: false,
HonorTimestamps: true,
HTTPClientConfig: config.DefaultHTTPClientConfig,
// ScrapeTimeout and ScrapeInterval default to the configured
// globals.
ScrapeClassicHistograms: false,
MetricsPath: "/metrics",
Scheme: "http",
HonorLabels: false,
HonorTimestamps: true,
HTTPClientConfig: config.DefaultHTTPClientConfig,
}
// DefaultAlertmanagerConfig is the default alertmanager configuration.
@ -176,16 +175,16 @@ var (
// DefaultQueueConfig is the default remote queue configuration.
DefaultQueueConfig = QueueConfig{
// With a maximum of 200 shards, assuming an average of 100ms remote write
// time and 500 samples per batch, we will be able to push 1M samples/s.
MaxShards: 200,
// With a maximum of 50 shards, assuming an average of 100ms remote write
// time and 2000 samples per batch, we will be able to push 1M samples/s.
MaxShards: 50,
MinShards: 1,
MaxSamplesPerSend: 500,
MaxSamplesPerSend: 2000,
// Each shard will have a max of 2500 samples pending in its channel, plus the pending
// samples that have been enqueued. Theoretically we should only ever have about 3000 samples
// per shard pending. At 200 shards that's 600k.
Capacity: 2500,
// Each shard will have a max of 10,000 samples pending in its channel, plus the pending
// samples that have been enqueued. Theoretically we should only ever have about 12,000 samples
// per shard pending. At 50 shards that's 600k.
Capacity: 10000,
BatchSendDeadline: model.Duration(5 * time.Second),
// Backoff times for retrying a batch of samples on recoverable errors.
@ -197,7 +196,7 @@ var (
DefaultMetadataConfig = MetadataConfig{
Send: true,
SendInterval: model.Duration(1 * time.Minute),
MaxSamplesPerSend: 500,
MaxSamplesPerSend: 2000,
}
// DefaultRemoteReadConfig is the default remote read configuration.
@ -219,12 +218,13 @@ var (
// Config is the top-level configuration for Prometheus's config files.
type Config struct {
GlobalConfig GlobalConfig `yaml:"global"`
AlertingConfig AlertingConfig `yaml:"alerting,omitempty"`
RuleFiles []string `yaml:"rule_files,omitempty"`
ScrapeConfigs []*ScrapeConfig `yaml:"scrape_configs,omitempty"`
StorageConfig StorageConfig `yaml:"storage,omitempty"`
TracingConfig TracingConfig `yaml:"tracing,omitempty"`
GlobalConfig GlobalConfig `yaml:"global"`
AlertingConfig AlertingConfig `yaml:"alerting,omitempty"`
RuleFiles []string `yaml:"rule_files,omitempty"`
ScrapeConfigFiles []string `yaml:"scrape_config_files,omitempty"`
ScrapeConfigs []*ScrapeConfig `yaml:"scrape_configs,omitempty"`
StorageConfig StorageConfig `yaml:"storage,omitempty"`
TracingConfig TracingConfig `yaml:"tracing,omitempty"`
RemoteWriteConfigs []*RemoteWriteConfig `yaml:"remote_write,omitempty"`
RemoteReadConfigs []*RemoteReadConfig `yaml:"remote_read,omitempty"`
@ -238,6 +238,9 @@ func (c *Config) SetDirectory(dir string) {
for i, file := range c.RuleFiles {
c.RuleFiles[i] = config.JoinDir(dir, file)
}
for i, file := range c.ScrapeConfigFiles {
c.ScrapeConfigFiles[i] = config.JoinDir(dir, file)
}
for _, c := range c.ScrapeConfigs {
c.SetDirectory(dir)
}
@ -257,6 +260,58 @@ func (c Config) String() string {
return string(b)
}
// ScrapeConfigs returns the scrape configurations.
func (c *Config) GetScrapeConfigs() ([]*ScrapeConfig, error) {
scfgs := make([]*ScrapeConfig, len(c.ScrapeConfigs))
jobNames := map[string]string{}
for i, scfg := range c.ScrapeConfigs {
// We do these checks for library users that would not call Validate in
// Unmarshal.
if err := scfg.Validate(c.GlobalConfig); err != nil {
return nil, err
}
if _, ok := jobNames[scfg.JobName]; ok {
return nil, fmt.Errorf("found multiple scrape configs with job name %q", scfg.JobName)
}
jobNames[scfg.JobName] = "main config file"
scfgs[i] = scfg
}
for _, pat := range c.ScrapeConfigFiles {
fs, err := filepath.Glob(pat)
if err != nil {
// The only error can be a bad pattern.
return nil, fmt.Errorf("error retrieving scrape config files for %q: %w", pat, err)
}
for _, filename := range fs {
cfg := ScrapeConfigs{}
content, err := os.ReadFile(filename)
if err != nil {
return nil, fileErr(filename, err)
}
err = yaml.UnmarshalStrict(content, &cfg)
if err != nil {
return nil, fileErr(filename, err)
}
for _, scfg := range cfg.ScrapeConfigs {
if err := scfg.Validate(c.GlobalConfig); err != nil {
return nil, fileErr(filename, err)
}
if f, ok := jobNames[scfg.JobName]; ok {
return nil, fileErr(filename, fmt.Errorf("found multiple scrape configs with job name %q, first found in %s", scfg.JobName, f))
}
jobNames[scfg.JobName] = fmt.Sprintf("%q", filePath(filename))
scfg.SetDirectory(filepath.Dir(filename))
scfgs = append(scfgs, scfg)
}
}
}
return scfgs, nil
}
// UnmarshalYAML implements the yaml.Unmarshaler interface.
func (c *Config) UnmarshalYAML(unmarshal func(interface{}) error) error {
*c = DefaultConfig
@ -279,26 +334,18 @@ func (c *Config) UnmarshalYAML(unmarshal func(interface{}) error) error {
return fmt.Errorf("invalid rule file path %q", rf)
}
}
for _, sf := range c.ScrapeConfigFiles {
if !patRulePath.MatchString(sf) {
return fmt.Errorf("invalid scrape config file path %q", sf)
}
}
// Do global overrides and validate unique names.
jobNames := map[string]struct{}{}
for _, scfg := range c.ScrapeConfigs {
if scfg == nil {
return errors.New("empty or null scrape config section")
}
// First set the correct scrape interval, then check that the timeout
// (inferred or explicit) is not greater than that.
if scfg.ScrapeInterval == 0 {
scfg.ScrapeInterval = c.GlobalConfig.ScrapeInterval
}
if scfg.ScrapeTimeout > scfg.ScrapeInterval {
return fmt.Errorf("scrape timeout greater than scrape interval for scrape config with job name %q", scfg.JobName)
}
if scfg.ScrapeTimeout == 0 {
if c.GlobalConfig.ScrapeTimeout > scfg.ScrapeInterval {
scfg.ScrapeTimeout = scfg.ScrapeInterval
} else {
scfg.ScrapeTimeout = c.GlobalConfig.ScrapeTimeout
}
if err := scfg.Validate(c.GlobalConfig); err != nil {
return err
}
if _, ok := jobNames[scfg.JobName]; ok {
@ -344,6 +391,24 @@ type GlobalConfig struct {
QueryLogFile string `yaml:"query_log_file,omitempty"`
// The labels to add to any timeseries that this Prometheus instance scrapes.
ExternalLabels labels.Labels `yaml:"external_labels,omitempty"`
// An uncompressed response body larger than this many bytes will cause the
// scrape to fail. 0 means no limit.
BodySizeLimit units.Base2Bytes `yaml:"body_size_limit,omitempty"`
// More than this many samples post metric-relabeling will cause the scrape to
// fail. 0 means no limit.
SampleLimit uint `yaml:"sample_limit,omitempty"`
// More than this many targets after the target relabeling will cause the
// scrapes to fail. 0 means no limit.
TargetLimit uint `yaml:"target_limit,omitempty"`
// More than this many labels post metric-relabeling will cause the scrape to
// fail. 0 means no limit.
LabelLimit uint `yaml:"label_limit,omitempty"`
// More than this label name length post metric-relabeling will cause the
// scrape to fail. 0 means no limit.
LabelNameLengthLimit uint `yaml:"label_name_length_limit,omitempty"`
// More than this label value length post metric-relabeling will cause the
// scrape to fail. 0 means no limit.
LabelValueLengthLimit uint `yaml:"label_value_length_limit,omitempty"`
}
// SetDirectory joins any relative file paths with dir.
@ -361,13 +426,16 @@ func (c *GlobalConfig) UnmarshalYAML(unmarshal func(interface{}) error) error {
return err
}
for _, l := range gc.ExternalLabels {
if err := gc.ExternalLabels.Validate(func(l labels.Label) error {
if !model.LabelName(l.Name).IsValid() {
return fmt.Errorf("%q is not a valid label name", l.Name)
}
if !model.LabelValue(l.Value).IsValid() {
return fmt.Errorf("%q is not a valid label value", l.Value)
}
return nil
}); err != nil {
return err
}
// First set the correct scrape interval, then check that the timeout
@ -394,13 +462,17 @@ func (c *GlobalConfig) UnmarshalYAML(unmarshal func(interface{}) error) error {
// isZero returns true iff the global config is the zero value.
func (c *GlobalConfig) isZero() bool {
return c.ExternalLabels == nil &&
return c.ExternalLabels.IsEmpty() &&
c.ScrapeInterval == 0 &&
c.ScrapeTimeout == 0 &&
c.EvaluationInterval == 0 &&
c.QueryLogFile == ""
}
type ScrapeConfigs struct {
ScrapeConfigs []*ScrapeConfig `yaml:"scrape_configs,omitempty"`
}
// ScrapeConfig configures a scraping unit for Prometheus.
type ScrapeConfig struct {
// The job name to which the job label is set by default.
@ -415,6 +487,8 @@ type ScrapeConfig struct {
ScrapeInterval model.Duration `yaml:"scrape_interval,omitempty"`
// The timeout for scraping targets of this config.
ScrapeTimeout model.Duration `yaml:"scrape_timeout,omitempty"`
// Whether to scrape a classic histogram that is also exposed as a native histogram.
ScrapeClassicHistograms bool `yaml:"scrape_classic_histograms,omitempty"`
// The HTTP resource path on which to fetch metrics from targets.
MetricsPath string `yaml:"metrics_path,omitempty"`
// The URL scheme with which to fetch metrics from targets.
@ -423,20 +497,23 @@ type ScrapeConfig struct {
// scrape to fail. 0 means no limit.
BodySizeLimit units.Base2Bytes `yaml:"body_size_limit,omitempty"`
// More than this many samples post metric-relabeling will cause the scrape to
// fail.
// fail. 0 means no limit.
SampleLimit uint `yaml:"sample_limit,omitempty"`
// More than this many targets after the target relabeling will cause the
// scrapes to fail.
// scrapes to fail. 0 means no limit.
TargetLimit uint `yaml:"target_limit,omitempty"`
// More than this many labels post metric-relabeling will cause the scrape to
// fail.
// fail. 0 means no limit.
LabelLimit uint `yaml:"label_limit,omitempty"`
// More than this label name length post metric-relabeling will cause the
// scrape to fail.
// scrape to fail. 0 means no limit.
LabelNameLengthLimit uint `yaml:"label_name_length_limit,omitempty"`
// More than this label value length post metric-relabeling will cause the
// scrape to fail.
// scrape to fail. 0 means no limit.
LabelValueLengthLimit uint `yaml:"label_value_length_limit,omitempty"`
// More than this many buckets in a native histogram will cause the scrape to
// fail.
NativeHistogramBucketLimit uint `yaml:"native_histogram_bucket_limit,omitempty"`
// We cannot do proper Go type embedding below as the parser will then parse
// values arbitrarily into the overflow maps of further-down types.
@ -494,6 +571,47 @@ func (c *ScrapeConfig) UnmarshalYAML(unmarshal func(interface{}) error) error {
return nil
}
func (c *ScrapeConfig) Validate(globalConfig GlobalConfig) error {
if c == nil {
return errors.New("empty or null scrape config section")
}
// First set the correct scrape interval, then check that the timeout
// (inferred or explicit) is not greater than that.
if c.ScrapeInterval == 0 {
c.ScrapeInterval = globalConfig.ScrapeInterval
}
if c.ScrapeTimeout > c.ScrapeInterval {
return fmt.Errorf("scrape timeout greater than scrape interval for scrape config with job name %q", c.JobName)
}
if c.ScrapeTimeout == 0 {
if globalConfig.ScrapeTimeout > c.ScrapeInterval {
c.ScrapeTimeout = c.ScrapeInterval
} else {
c.ScrapeTimeout = globalConfig.ScrapeTimeout
}
}
if c.BodySizeLimit == 0 {
c.BodySizeLimit = globalConfig.BodySizeLimit
}
if c.SampleLimit == 0 {
c.SampleLimit = globalConfig.SampleLimit
}
if c.TargetLimit == 0 {
c.TargetLimit = globalConfig.TargetLimit
}
if c.LabelLimit == 0 {
c.LabelLimit = globalConfig.LabelLimit
}
if c.LabelNameLengthLimit == 0 {
c.LabelNameLengthLimit = globalConfig.LabelNameLengthLimit
}
if c.LabelValueLengthLimit == 0 {
c.LabelValueLengthLimit = globalConfig.LabelValueLengthLimit
}
return nil
}
// MarshalYAML implements the yaml.Marshaler interface.
func (c *ScrapeConfig) MarshalYAML() (interface{}, error) {
return discovery.MarshalYAMLWithInlineConfigs(c)
@ -776,12 +894,13 @@ func CheckTargetAddress(address model.LabelValue) error {
// RemoteWriteConfig is the configuration for writing to remote storage.
type RemoteWriteConfig struct {
URL *config.URL `yaml:"url"`
RemoteTimeout model.Duration `yaml:"remote_timeout,omitempty"`
Headers map[string]string `yaml:"headers,omitempty"`
WriteRelabelConfigs []*relabel.Config `yaml:"write_relabel_configs,omitempty"`
Name string `yaml:"name,omitempty"`
SendExemplars bool `yaml:"send_exemplars,omitempty"`
URL *config.URL `yaml:"url"`
RemoteTimeout model.Duration `yaml:"remote_timeout,omitempty"`
Headers map[string]string `yaml:"headers,omitempty"`
WriteRelabelConfigs []*relabel.Config `yaml:"write_relabel_configs,omitempty"`
Name string `yaml:"name,omitempty"`
SendExemplars bool `yaml:"send_exemplars,omitempty"`
SendNativeHistograms bool `yaml:"send_native_histograms,omitempty"`
// We cannot do proper Go type embedding below as the parser will then parse
// values arbitrarily into the overflow maps of further-down types.
@ -789,6 +908,7 @@ type RemoteWriteConfig struct {
QueueConfig QueueConfig `yaml:"queue_config,omitempty"`
MetadataConfig MetadataConfig `yaml:"metadata_config,omitempty"`
SigV4Config *sigv4.SigV4Config `yaml:"sigv4,omitempty"`
AzureADConfig *azuread.AzureADConfig `yaml:"azuread,omitempty"`
}
// SetDirectory joins any relative file paths with dir.
@ -825,8 +945,12 @@ func (c *RemoteWriteConfig) UnmarshalYAML(unmarshal func(interface{}) error) err
httpClientConfigAuthEnabled := c.HTTPClientConfig.BasicAuth != nil ||
c.HTTPClientConfig.Authorization != nil || c.HTTPClientConfig.OAuth2 != nil
if httpClientConfigAuthEnabled && c.SigV4Config != nil {
return fmt.Errorf("at most one of basic_auth, authorization, oauth2, & sigv4 must be configured")
if httpClientConfigAuthEnabled && (c.SigV4Config != nil || c.AzureADConfig != nil) {
return fmt.Errorf("at most one of basic_auth, authorization, oauth2, sigv4, & azuread must be configured")
}
if c.SigV4Config != nil && c.AzureADConfig != nil {
return fmt.Errorf("at most one of basic_auth, authorization, oauth2, sigv4, & azuread must be configured")
}
return nil
@ -847,7 +971,7 @@ func validateHeadersForTracing(headers map[string]string) error {
func validateHeaders(headers map[string]string) error {
for header := range headers {
if strings.ToLower(header) == "authorization" {
return errors.New("authorization header must be changed via the basic_auth, authorization, oauth2, or sigv4 parameter")
return errors.New("authorization header must be changed via the basic_auth, authorization, oauth2, sigv4, or azuread parameter")
}
if _, ok := reservedHeaders[strings.ToLower(header)]; ok {
return fmt.Errorf("%s is a reserved header. It must not be changed", header)
@ -935,3 +1059,15 @@ func (c *RemoteReadConfig) UnmarshalYAML(unmarshal func(interface{}) error) erro
// Thus we just do its validation here.
return c.HTTPClientConfig.Validate()
}
func filePath(filename string) string {
absPath, err := filepath.Abs(filename)
if err != nil {
return filename
}
return absPath
}
func fileErr(filename string, err error) error {
return fmt.Errorf("%q: %w", filePath(filename), err)
}

672
config/config_test.go
View file

@ -47,6 +47,7 @@ import (
"github.com/prometheus/prometheus/discovery/moby"
"github.com/prometheus/prometheus/discovery/nomad"
"github.com/prometheus/prometheus/discovery/openstack"
"github.com/prometheus/prometheus/discovery/ovhcloud"
"github.com/prometheus/prometheus/discovery/puppetdb"
"github.com/prometheus/prometheus/discovery/scaleway"
"github.com/prometheus/prometheus/discovery/targetgroup"
@ -67,6 +68,15 @@ func mustParseURL(u string) *config.URL {
return &config.URL{URL: parsed}
}
const (
globBodySizeLimit = 15 * units.MiB
globSampleLimit = 1500
globTargetLimit = 30
globLabelLimit = 30
globLabelNameLengthLimit = 200
globLabelValueLengthLimit = 200
)
var expectedConf = &Config{
GlobalConfig: GlobalConfig{
ScrapeInterval: model.Duration(15 * time.Second),
@ -75,6 +85,13 @@ var expectedConf = &Config{
QueryLogFile: "",
ExternalLabels: labels.FromStrings("foo", "bar", "monitor", "codelab"),
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
},
RuleFiles: []string{
@ -164,10 +181,16 @@ var expectedConf = &Config{
{
JobName: "prometheus",
HonorLabels: true,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorLabels: true,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -216,36 +239,59 @@ var expectedConf = &Config{
Regex: relabel.MustNewRegexp("(.*)some-[regex]"),
Replacement: "foo-${1}",
Action: relabel.Replace,
}, {
},
{
SourceLabels: model.LabelNames{"abc"},
TargetLabel: "cde",
Separator: ";",
Regex: relabel.DefaultRelabelConfig.Regex,
Replacement: relabel.DefaultRelabelConfig.Replacement,
Action: relabel.Replace,
}, {
},
{
TargetLabel: "abc",
Separator: ";",
Regex: relabel.DefaultRelabelConfig.Regex,
Replacement: "static",
Action: relabel.Replace,
}, {
},
{
TargetLabel: "abc",
Separator: ";",
Regex: relabel.MustNewRegexp(""),
Replacement: "static",
Action: relabel.Replace,
},
{
SourceLabels: model.LabelNames{"foo"},
TargetLabel: "abc",
Action: relabel.KeepEqual,
Regex: relabel.DefaultRelabelConfig.Regex,
Replacement: relabel.DefaultRelabelConfig.Replacement,
Separator: relabel.DefaultRelabelConfig.Separator,
},
{
SourceLabels: model.LabelNames{"foo"},
TargetLabel: "abc",
Action: relabel.DropEqual,
Regex: relabel.DefaultRelabelConfig.Regex,
Replacement: relabel.DefaultRelabelConfig.Replacement,
Separator: relabel.DefaultRelabelConfig.Separator,
},
},
},
{
JobName: "service-x",
HonorTimestamps: true,
ScrapeInterval: model.Duration(50 * time.Second),
ScrapeTimeout: model.Duration(5 * time.Second),
BodySizeLimit: 10 * units.MiB,
SampleLimit: 1000,
HonorTimestamps: true,
ScrapeInterval: model.Duration(50 * time.Second),
ScrapeTimeout: model.Duration(5 * time.Second),
BodySizeLimit: 10 * units.MiB,
SampleLimit: 1000,
TargetLimit: 35,
LabelLimit: 35,
LabelNameLengthLimit: 210,
LabelValueLengthLimit: 210,
HTTPClientConfig: config.HTTPClientConfig{
BasicAuth: &config.BasicAuth{
@ -332,9 +378,15 @@ var expectedConf = &Config{
{
JobName: "service-y",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -343,6 +395,7 @@ var expectedConf = &Config{
ServiceDiscoveryConfigs: discovery.Configs{
&consul.SDConfig{
Server: "localhost:1234",
PathPrefix: "/consul",
Token: "mysecret",
Services: []string{"nginx", "cache", "mysql"},
ServiceTags: []string{"canary", "v1"},
@ -378,9 +431,15 @@ var expectedConf = &Config{
{
JobName: "service-z",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: model.Duration(10 * time.Second),
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: model.Duration(10 * time.Second),
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: "/metrics",
Scheme: "http",
@ -403,9 +462,15 @@ var expectedConf = &Config{
{
JobName: "service-kubernetes",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -434,9 +499,15 @@ var expectedConf = &Config{
{
JobName: "service-kubernetes-namespaces",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -465,9 +536,15 @@ var expectedConf = &Config{
{
JobName: "service-kuma",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -485,9 +562,15 @@ var expectedConf = &Config{
{
JobName: "service-marathon",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -514,9 +597,15 @@ var expectedConf = &Config{
{
JobName: "service-nomad",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -540,9 +629,15 @@ var expectedConf = &Config{
{
JobName: "service-ec2",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -566,15 +661,22 @@ var expectedConf = &Config{
Values: []string{"web", "db"},
},
},
HTTPClientConfig: config.DefaultHTTPClientConfig,
},
},
},
{
JobName: "service-lightsail",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -582,21 +684,28 @@ var expectedConf = &Config{
ServiceDiscoveryConfigs: discovery.Configs{
&aws.LightsailSDConfig{
Region: "us-east-1",
AccessKey: "access",
SecretKey: "mysecret",
Profile: "profile",
RefreshInterval: model.Duration(60 * time.Second),
Port: 80,
Region: "us-east-1",
AccessKey: "access",
SecretKey: "mysecret",
Profile: "profile",
RefreshInterval: model.Duration(60 * time.Second),
Port: 80,
HTTPClientConfig: config.DefaultHTTPClientConfig,
},
},
},
{
JobName: "service-azure",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -620,9 +729,15 @@ var expectedConf = &Config{
{
JobName: "service-nerve",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -639,9 +754,15 @@ var expectedConf = &Config{
{
JobName: "0123service-xxx",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -661,9 +782,15 @@ var expectedConf = &Config{
{
JobName: "badfederation",
HonorTimestamps: false,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: false,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: "/federate",
Scheme: DefaultScrapeConfig.Scheme,
@ -683,9 +810,15 @@ var expectedConf = &Config{
{
JobName: "測試",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -705,9 +838,15 @@ var expectedConf = &Config{
{
JobName: "httpsd",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -724,9 +863,15 @@ var expectedConf = &Config{
{
JobName: "service-triton",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -751,9 +896,15 @@ var expectedConf = &Config{
{
JobName: "digitalocean-droplets",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -777,9 +928,15 @@ var expectedConf = &Config{
{
JobName: "docker",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -799,9 +956,15 @@ var expectedConf = &Config{
{
JobName: "dockerswarm",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -821,9 +984,15 @@ var expectedConf = &Config{
{
JobName: "service-openstack",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -847,9 +1016,15 @@ var expectedConf = &Config{
{
JobName: "service-puppetdb",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -875,10 +1050,16 @@ var expectedConf = &Config{
},
},
{
JobName: "hetzner",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
JobName: "hetzner",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -924,9 +1105,15 @@ var expectedConf = &Config{
{
JobName: "service-eureka",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -940,12 +1127,55 @@ var expectedConf = &Config{
},
},
},
{
JobName: "ovhcloud",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
HTTPClientConfig: config.DefaultHTTPClientConfig,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
ServiceDiscoveryConfigs: discovery.Configs{
&ovhcloud.SDConfig{
Endpoint: "ovh-eu",
ApplicationKey: "testAppKey",
ApplicationSecret: "testAppSecret",
ConsumerKey: "testConsumerKey",
RefreshInterval: model.Duration(60 * time.Second),
Service: "vps",
},
&ovhcloud.SDConfig{
Endpoint: "ovh-eu",
ApplicationKey: "testAppKey",
ApplicationSecret: "testAppSecret",
ConsumerKey: "testConsumerKey",
RefreshInterval: model.Duration(60 * time.Second),
Service: "dedicated_server",
},
},
},
{
JobName: "scaleway",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
HTTPClientConfig: config.DefaultHTTPClientConfig,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -978,9 +1208,15 @@ var expectedConf = &Config{
{
JobName: "linode-instances",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -1005,9 +1241,16 @@ var expectedConf = &Config{
{
JobName: "uyuni",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
HTTPClientConfig: config.DefaultHTTPClientConfig,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -1024,10 +1267,16 @@ var expectedConf = &Config{
},
},
{
JobName: "ionos",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
JobName: "ionos",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -1049,9 +1298,15 @@ var expectedConf = &Config{
{
JobName: "vultr",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
BodySizeLimit: globBodySizeLimit,
SampleLimit: globSampleLimit,
TargetLimit: globTargetLimit,
LabelLimit: globLabelLimit,
LabelNameLengthLimit: globLabelNameLengthLimit,
LabelValueLengthLimit: globLabelValueLengthLimit,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
@ -1175,7 +1430,7 @@ func TestElideSecrets(t *testing.T) {
yamlConfig := string(config)
matches := secretRe.FindAllStringIndex(yamlConfig, -1)
require.Equal(t, 18, len(matches), "wrong number of secret matches found")
require.Equal(t, 22, len(matches), "wrong number of secret matches found")
require.NotContains(t, yamlConfig, "mysecret",
"yaml marshal reveals authentication credentials.")
}
@ -1286,6 +1541,22 @@ var expectedErrors = []struct {
filename: "labeldrop5.bad.yml",
errMsg: "labeldrop action requires only 'regex', and no other fields",
},
{
filename: "dropequal.bad.yml",
errMsg: "relabel configuration for dropequal action requires 'target_label' value",
},
{
filename: "dropequal1.bad.yml",
errMsg: "dropequal action requires only 'source_labels' and `target_label`, and no other fields",
},
{
filename: "keepequal.bad.yml",
errMsg: "relabel configuration for keepequal action requires 'target_label' value",
},
{
filename: "keepequal1.bad.yml",
errMsg: "keepequal action requires only 'source_labels' and `target_label`, and no other fields",
},
{
filename: "labelmap.bad.yml",
errMsg: "\"l-$1\" is invalid 'replacement' for labelmap action",
@ -1456,7 +1727,7 @@ var expectedErrors = []struct {
},
{
filename: "remote_write_authorization_header.bad.yml",
errMsg: `authorization header must be changed via the basic_auth, authorization, oauth2, or sigv4 parameter`,
errMsg: `authorization header must be changed via the basic_auth, authorization, oauth2, sigv4, or azuread parameter`,
},
{
filename: "remote_write_url_missing.bad.yml",
@ -1618,6 +1889,18 @@ var expectedErrors = []struct {
filename: "ionos_datacenter.bad.yml",
errMsg: "datacenter id can't be empty",
},
{
filename: "ovhcloud_no_secret.bad.yml",
errMsg: "application secret can not be empty",
},
{
filename: "ovhcloud_bad_service.bad.yml",
errMsg: "unknown service: fakeservice",
},
{
filename: "scrape_config_files_glob.bad.yml",
errMsg: `parsing YAML file testdata/scrape_config_files_glob.bad.yml: invalid scrape config file path "scrape_configs/*/*"`,
},
}
func TestBadConfigs(t *testing.T) {
@ -1670,6 +1953,33 @@ func TestExpandExternalLabels(t *testing.T) {
require.Equal(t, labels.FromStrings("bar", "foo", "baz", "fooTestValuebar", "foo", "TestValue", "qux", "foo${TEST}", "xyz", "foo$bar"), c.GlobalConfig.ExternalLabels)
}
func TestAgentMode(t *testing.T) {
_, err := LoadFile("testdata/agent_mode.with_alert_manager.yml", true, false, log.NewNopLogger())
require.ErrorContains(t, err, "field alerting is not allowed in agent mode")
_, err = LoadFile("testdata/agent_mode.with_alert_relabels.yml", true, false, log.NewNopLogger())
require.ErrorContains(t, err, "field alerting is not allowed in agent mode")
_, err = LoadFile("testdata/agent_mode.with_rule_files.yml", true, false, log.NewNopLogger())
require.ErrorContains(t, err, "field rule_files is not allowed in agent mode")
_, err = LoadFile("testdata/agent_mode.with_remote_reads.yml", true, false, log.NewNopLogger())
require.ErrorContains(t, err, "field remote_read is not allowed in agent mode")
c, err := LoadFile("testdata/agent_mode.without_remote_writes.yml", true, false, log.NewNopLogger())
require.NoError(t, err)
require.Len(t, c.RemoteWriteConfigs, 0)
c, err = LoadFile("testdata/agent_mode.good.yml", true, false, log.NewNopLogger())
require.NoError(t, err)
require.Len(t, c.RemoteWriteConfigs, 1)
require.Equal(
t,
"http://remote1/push",
c.RemoteWriteConfigs[0].URL.String(),
)
}
func TestEmptyGlobalBlock(t *testing.T) {
c, err := Load("global:\n", false, log.NewNopLogger())
require.NoError(t, err)
@ -1677,6 +1987,156 @@ func TestEmptyGlobalBlock(t *testing.T) {
require.Equal(t, exp, *c)
}
func TestGetScrapeConfigs(t *testing.T) {
sc := func(jobName string, scrapeInterval, scrapeTimeout model.Duration) *ScrapeConfig {
return &ScrapeConfig{
JobName: jobName,
HonorTimestamps: true,
ScrapeInterval: scrapeInterval,
ScrapeTimeout: scrapeTimeout,
MetricsPath: "/metrics",
Scheme: "http",
HTTPClientConfig: config.DefaultHTTPClientConfig,
ServiceDiscoveryConfigs: discovery.Configs{
discovery.StaticConfig{
{
Targets: []model.LabelSet{
{
model.AddressLabel: "localhost:8080",
},
},
Source: "0",
},
},
},
}
}
testCases := []struct {
name string
configFile string
expectedResult []*ScrapeConfig
expectedError string
}{
{
name: "An included config file should be a valid global config.",
configFile: "testdata/scrape_config_files.good.yml",
expectedResult: []*ScrapeConfig{sc("prometheus", model.Duration(60*time.Second), model.Duration(10*time.Second))},
},
{
name: "An global config that only include a scrape config file.",
configFile: "testdata/scrape_config_files_only.good.yml",
expectedResult: []*ScrapeConfig{sc("prometheus", model.Duration(60*time.Second), model.Duration(10*time.Second))},
},
{
name: "An global config that combine scrape config files and scrape configs.",
configFile: "testdata/scrape_config_files_combined.good.yml",
expectedResult: []*ScrapeConfig{
sc("node", model.Duration(60*time.Second), model.Duration(10*time.Second)),
sc("prometheus", model.Duration(60*time.Second), model.Duration(10*time.Second)),
sc("alertmanager", model.Duration(60*time.Second), model.Duration(10*time.Second)),
},
},
{
name: "An global config that includes a scrape config file with globs",
configFile: "testdata/scrape_config_files_glob.good.yml",
expectedResult: []*ScrapeConfig{
{
JobName: "prometheus",
HonorTimestamps: true,
ScrapeInterval: model.Duration(60 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
HTTPClientConfig: config.HTTPClientConfig{
TLSConfig: config.TLSConfig{
CertFile: filepath.FromSlash("testdata/scrape_configs/valid_cert_file"),
KeyFile: filepath.FromSlash("testdata/scrape_configs/valid_key_file"),
},
FollowRedirects: true,
EnableHTTP2: true,
},
ServiceDiscoveryConfigs: discovery.Configs{
discovery.StaticConfig{
{
Targets: []model.LabelSet{
{model.AddressLabel: "localhost:8080"},
},
Source: "0",
},
},
},
},
{
JobName: "node",
HonorTimestamps: true,
ScrapeInterval: model.Duration(15 * time.Second),
ScrapeTimeout: DefaultGlobalConfig.ScrapeTimeout,
HTTPClientConfig: config.HTTPClientConfig{
TLSConfig: config.TLSConfig{
CertFile: filepath.FromSlash("testdata/valid_cert_file"),
KeyFile: filepath.FromSlash("testdata/valid_key_file"),
},
FollowRedirects: true,
EnableHTTP2: true,
},
MetricsPath: DefaultScrapeConfig.MetricsPath,
Scheme: DefaultScrapeConfig.Scheme,
ServiceDiscoveryConfigs: discovery.Configs{
&vultr.SDConfig{
HTTPClientConfig: config.HTTPClientConfig{
Authorization: &config.Authorization{
Type: "Bearer",
Credentials: "abcdef",
},
FollowRedirects: true,
EnableHTTP2: true,
},
Port: 80,
RefreshInterval: model.Duration(60 * time.Second),
},
},
},
},
},
{
name: "An global config that includes twice the same scrape configs.",
configFile: "testdata/scrape_config_files_double_import.bad.yml",
expectedError: `found multiple scrape configs with job name "prometheus"`,
},
{
name: "An global config that includes a scrape config identical to a scrape config in the main file.",
configFile: "testdata/scrape_config_files_duplicate.bad.yml",
expectedError: `found multiple scrape configs with job name "prometheus"`,
},
{
name: "An global config that includes a scrape config file with errors.",
configFile: "testdata/scrape_config_files_global.bad.yml",
expectedError: `scrape timeout greater than scrape interval for scrape config with job name "prometheus"`,
},
}
for _, tc := range testCases {
t.Run(tc.name, func(t *testing.T) {
c, err := LoadFile(tc.configFile, false, false, log.NewNopLogger())
require.NoError(t, err)
scfgs, err := c.GetScrapeConfigs()
if len(tc.expectedError) > 0 {
require.ErrorContains(t, err, tc.expectedError)
}
require.Equal(t, tc.expectedResult, scfgs)
})
}
}
func kubernetesSDHostURL() config.URL {
tURL, _ := url.Parse("https://localhost:1234")
return config.URL{URL: tURL}

2
config/testdata/agent_mode.good.yml vendored Normal file
View file

@ -0,0 +1,2 @@
remote_write:
- url: http://remote1/push

6
config/testdata/agent_mode.with_alert_manager.yml vendored Normal file
View file

@ -0,0 +1,6 @@
alerting:
alertmanagers:
- scheme: https
static_configs:
- targets:
- "1.2.3.4:9093"

5
config/testdata/agent_mode.with_alert_relabels.yml vendored Normal file
View file

@ -0,0 +1,5 @@
alerting:
alert_relabel_configs:
- action: uppercase
source_labels: [instance]
target_label: instance

5
config/testdata/agent_mode.with_remote_reads.yml vendored Normal file
View file

@ -0,0 +1,5 @@
remote_read:
- url: http://remote1/read
read_recent: true
name: default
enable_http2: false

3
config/testdata/agent_mode.with_rule_files.yml vendored Normal file
View file

@ -0,0 +1,3 @@
rule_files:
- "first.rules"
- "my/*.rules"

2
config/testdata/agent_mode.without_remote_writes.yml vendored Normal file
View file

@ -0,0 +1,2 @@
global:
scrape_interval: 15s

33
config/testdata/conf.good.yml vendored
View file

@ -2,6 +2,12 @@
global:
scrape_interval: 15s
evaluation_interval: 30s
body_size_limit: 15MB
sample_limit: 1500
target_limit: 30
label_limit: 30
label_name_length_limit: 200
label_value_length_limit: 200
# scrape_timeout is set to the global default (10s).
external_labels:
@ -87,6 +93,12 @@ scrape_configs:
- regex:
replacement: static
target_label: abc
- source_labels: [foo]
target_label: abc
action: keepequal
- source_labels: [foo]
target_label: abc
action: dropequal
authorization:
credentials_file: valid_token_file
@ -105,6 +117,11 @@ scrape_configs:
body_size_limit: 10MB
sample_limit: 1000
target_limit: 35
label_limit: 35
label_name_length_limit: 210
label_value_length_limit: 210
metrics_path: /my_path
scheme: https
@ -145,6 +162,7 @@ scrape_configs:
consul_sd_configs:
- server: "localhost:1234"
token: mysecret
path_prefix: /consul
services: ["nginx", "cache", "mysql"]
tags: ["canary", "v1"]
node_meta:
@ -349,6 +367,21 @@ scrape_configs:
eureka_sd_configs:
- server: "http://eureka.example.com:8761/eureka"
- job_name: ovhcloud
ovhcloud_sd_configs:
- service: vps
endpoint: ovh-eu
application_key: testAppKey
application_secret: testAppSecret
consumer_key: testConsumerKey
refresh_interval: 1m
- service: dedicated_server
endpoint: ovh-eu
application_key: testAppKey
application_secret: testAppSecret
consumer_key: testConsumerKey
refresh_interval: 1m
- job_name: scaleway
scaleway_sd_configs:
- role: instance

5
config/testdata/dropequal.bad.yml vendored Normal file
View file

@ -0,0 +1,5 @@
scrape_configs:
- job_name: prometheus
relabel_configs:
- source_labels: [abcdef]
action: dropequal

7
config/testdata/dropequal1.bad.yml vendored Normal file
View file

@ -0,0 +1,7 @@
scrape_configs:
- job_name: prometheus
relabel_configs:
- source_labels: [abcdef]
action: dropequal
regex: foo
target_label: bar

5
config/testdata/keepequal.bad.yml vendored Normal file
View file

@ -0,0 +1,5 @@
scrape_configs:
- job_name: prometheus
relabel_configs:
- source_labels: [abcdef]
action: keepequal

7
config/testdata/keepequal1.bad.yml vendored Normal file
View file

@ -0,0 +1,7 @@
scrape_configs:
- job_name: prometheus
relabel_configs:
- source_labels: [abcdef]
action: keepequal
regex: foo
target_label: bar

8
config/testdata/ovhcloud_bad_service.bad.yml vendored Normal file
View file

@ -0,0 +1,8 @@
scrape_configs:
- ovhcloud_sd_configs:
- service: fakeservice
endpoint: ovh-eu
application_key: testAppKey
application_secret: testAppSecret
consumer_key: testConsumerKey
refresh_interval: 1m

7
config/testdata/ovhcloud_no_secret.bad.yml vendored Normal file
View file

@ -0,0 +1,7 @@
scrape_configs:
- ovhcloud_sd_configs:
- service: dedicated_server
endpoint: ovh-eu
application_key: testAppKey
consumer_key: testConsumerKey
refresh_interval: 1m

6
config/testdata/scrape_config_files.bad.yml vendored Normal file
View file

@ -0,0 +1,6 @@
scrape_configs:
- job_name: prometheus
scrape_interval: 10s
scrape_timeout: 20s
static_configs:
- targets: ['localhost:8080']

4
config/testdata/scrape_config_files.good.yml vendored Normal file
View file

@ -0,0 +1,4 @@
scrape_configs:
- job_name: prometheus
static_configs:
- targets: ['localhost:8080']

4
config/testdata/scrape_config_files2.good.yml vendored Normal file
View file

@ -0,0 +1,4 @@
scrape_configs:
- job_name: alertmanager
static_configs:
- targets: ['localhost:8080']

7
config/testdata/scrape_config_files_combined.good.yml vendored Normal file
View file

@ -0,0 +1,7 @@
scrape_config_files:
- scrape_config_files.good.yml
- scrape_config_files2.good.yml
scrape_configs:
- job_name: node
static_configs:
- targets: ['localhost:8080']

3
config/testdata/scrape_config_files_double_import.bad.yml vendored Normal file
View file

@ -0,0 +1,3 @@
scrape_config_files:
- scrape_config_files.good.yml
- scrape_config_files.good.yml

6
config/testdata/scrape_config_files_duplicate.bad.yml vendored Normal file
View file

@ -0,0 +1,6 @@
scrape_config_files:
- scrape_config_files.good.yml
scrape_configs:
- job_name: prometheus
static_configs:
- targets: ['localhost:8080']

6
config/testdata/scrape_config_files_glob.bad.yml vendored Normal file
View file

@ -0,0 +1,6 @@
scrape_config_files:
- scrape_configs/*/*
scrape_configs:
- job_name: node
static_configs:
- targets: ['localhost:8080']

2
config/testdata/scrape_config_files_glob.good.yml vendored Normal file
View file

@ -0,0 +1,2 @@
scrape_config_files:
- scrape_configs/*.yml

2
config/testdata/scrape_config_files_global.bad.yml vendored Normal file
View file

@ -0,0 +1,2 @@
scrape_config_files:
- scrape_config_files.bad.yml

11
config/testdata/scrape_config_files_global_duplicate.bad.yml vendored Normal file
View file

@ -0,0 +1,11 @@
global:
scrape_interval: 15s
scrape_config_files:
- scrape_config_files.good.yml
- scrape_config_files.good.yml
scrape_configs:
- job_name: prometheus
static_configs:
- targets: ['localhost:8080']

2
config/testdata/scrape_config_files_only.good.yml vendored Normal file
View file

@ -0,0 +1,2 @@
scrape_config_files:
- scrape_config_files.good.yml

7
config/testdata/scrape_configs/scrape_config_files1.good.yml vendored Normal file
View file

@ -0,0 +1,7 @@
scrape_configs:
- job_name: prometheus
static_configs:
- targets: ['localhost:8080']
tls_config:
cert_file: valid_cert_file
key_file: valid_key_file

9
config/testdata/scrape_configs/scrape_config_files2.good.yml vendored Normal file
View file

@ -0,0 +1,9 @@
scrape_configs:
- job_name: node
scrape_interval: 15s
tls_config:
cert_file: ../valid_cert_file
key_file: ../valid_key_file
vultr_sd_configs:
- authorization:
credentials: abcdef

15
discovery/aws/ec2.go
View file

@ -66,8 +66,9 @@ const (
// DefaultEC2SDConfig is the default EC2 SD configuration.
var DefaultEC2SDConfig = EC2SDConfig{
Port: 80,
RefreshInterval: model.Duration(60 * time.Second),
Port: 80,
RefreshInterval: model.Duration(60 * time.Second),
HTTPClientConfig: config.DefaultHTTPClientConfig,
}
func init() {
@ -91,6 +92,8 @@ type EC2SDConfig struct {
RefreshInterval model.Duration `yaml:"refresh_interval,omitempty"`
Port int `yaml:"port"`
Filters []*EC2Filter `yaml:"filters"`
HTTPClientConfig config.HTTPClientConfig `yaml:",inline"`
}
// Name returns the name of the EC2 Config.
@ -161,7 +164,7 @@ func NewEC2Discovery(conf *EC2SDConfig, logger log.Logger) *EC2Discovery {
return d
}
func (d *EC2Discovery) ec2Client(ctx context.Context) (*ec2.EC2, error) {
func (d *EC2Discovery) ec2Client(context.Context) (*ec2.EC2, error) {
if d.ec2 != nil {
return d.ec2, nil
}
@ -171,11 +174,17 @@ func (d *EC2Discovery) ec2Client(ctx context.Context) (*ec2.EC2, error) {
creds = nil
}
client, err := config.NewClientFromConfig(d.cfg.HTTPClientConfig, "ec2_sd")
if err != nil {
return nil, err
}
sess, err := session.NewSessionWithOptions(session.Options{
Config: aws.Config{
Endpoint: &d.cfg.Endpoint,
Region: &d.cfg.Region,
Credentials: creds,
HTTPClient: client,
},
Profile: d.cfg.Profile,
})

13
discovery/aws/lightsail.go
View file

@ -56,8 +56,9 @@ const (
// DefaultLightsailSDConfig is the default Lightsail SD configuration.
var DefaultLightsailSDConfig = LightsailSDConfig{
Port: 80,
RefreshInterval: model.Duration(60 * time.Second),
Port: 80,
RefreshInterval: model.Duration(60 * time.Second),
HTTPClientConfig: config.DefaultHTTPClientConfig,
}
func init() {
@ -74,6 +75,8 @@ type LightsailSDConfig struct {
RoleARN string `yaml:"role_arn,omitempty"`
RefreshInterval model.Duration `yaml:"refresh_interval,omitempty"`
Port int `yaml:"port"`
HTTPClientConfig config.HTTPClientConfig `yaml:",inline"`
}
// Name returns the name of the Lightsail Config.
@ -144,11 +147,17 @@ func (d *LightsailDiscovery) lightsailClient() (*lightsail.Lightsail, error) {
creds = nil
}
client, err := config.NewClientFromConfig(d.cfg.HTTPClientConfig, "lightsail_sd")
if err != nil {
return nil, err
}
sess, err := session.NewSessionWithOptions(session.Options{
Config: aws.Config{
Endpoint: &d.cfg.Endpoint,
Region: &d.cfg.Region,
Credentials: creds,
HTTPClient: client,
},
Profile: d.cfg.Profile,
})

27
discovery/azure/azure.go
View file

@ -55,6 +55,7 @@ const (
azureLabelMachinePublicIP = azureLabel + "machine_public_ip"
azureLabelMachineTag = azureLabel + "machine_tag_"
azureLabelMachineScaleSet = azureLabel + "machine_scale_set"
azureLabelMachineSize = azureLabel + "machine_size"
authMethodOAuth = "OAuth"
authMethodManagedIdentity = "ManagedIdentity"
@ -261,6 +262,7 @@ type virtualMachine struct {
ScaleSet string
Tags map[string]*string
NetworkInterfaces []string
Size string
}
// Create a new azureResource object from an ID string.
@ -343,6 +345,7 @@ func (d *Discovery) refresh(ctx context.Context) ([]*targetgroup.Group, error) {
azureLabelMachineOSType: model.LabelValue(vm.OsType),
azureLabelMachineLocation: model.LabelValue(vm.Location),
azureLabelMachineResourceGroup: model.LabelValue(r.ResourceGroup),
azureLabelMachineSize: model.LabelValue(vm.Size),
}
if vm.ScaleSet != "" {
@ -514,6 +517,7 @@ func mapFromVM(vm compute.VirtualMachine) virtualMachine {
tags := map[string]*string{}
networkInterfaces := []string{}
var computerName string
var size string
if vm.Tags != nil {
tags = vm.Tags
@ -525,10 +529,13 @@ func mapFromVM(vm compute.VirtualMachine) virtualMachine {
}
}
if vm.VirtualMachineProperties != nil &&
vm.VirtualMachineProperties.OsProfile != nil &&
vm.VirtualMachineProperties.OsProfile.ComputerName != nil {
computerName = *(vm.VirtualMachineProperties.OsProfile.ComputerName)
if vm.VirtualMachineProperties != nil {
if vm.VirtualMachineProperties.OsProfile != nil && vm.VirtualMachineProperties.OsProfile.ComputerName != nil {
computerName = *(vm.VirtualMachineProperties.OsProfile.ComputerName)
}
if vm.VirtualMachineProperties.HardwareProfile != nil {
size = string(vm.VirtualMachineProperties.HardwareProfile.VMSize)
}
}
return virtualMachine{
@ -541,6 +548,7 @@ func mapFromVM(vm compute.VirtualMachine) virtualMachine {
ScaleSet: "",
Tags: tags,
NetworkInterfaces: networkInterfaces,
Size: size,
}
}
@ -549,6 +557,7 @@ func mapFromVMScaleSetVM(vm compute.VirtualMachineScaleSetVM, scaleSetName strin
tags := map[string]*string{}
networkInterfaces := []string{}
var computerName string
var size string
if vm.Tags != nil {
tags = vm.Tags
@ -560,8 +569,13 @@ func mapFromVMScaleSetVM(vm compute.VirtualMachineScaleSetVM, scaleSetName strin
}
}
if vm.VirtualMachineScaleSetVMProperties != nil && vm.VirtualMachineScaleSetVMProperties.OsProfile != nil {
computerName = *(vm.VirtualMachineScaleSetVMProperties.OsProfile.ComputerName)
if vm.VirtualMachineScaleSetVMProperties != nil {
if vm.VirtualMachineScaleSetVMProperties.OsProfile != nil && vm.VirtualMachineScaleSetVMProperties.OsProfile.ComputerName != nil {
computerName = *(vm.VirtualMachineScaleSetVMProperties.OsProfile.ComputerName)
}
if vm.VirtualMachineScaleSetVMProperties.HardwareProfile != nil {
size = string(vm.VirtualMachineScaleSetVMProperties.HardwareProfile.VMSize)
}
}
return virtualMachine{
@ -574,6 +588,7 @@ func mapFromVMScaleSetVM(vm compute.VirtualMachineScaleSetVM, scaleSetName strin
ScaleSet: scaleSetName,
Tags: tags,
NetworkInterfaces: networkInterfaces,
Size: size,
}
}

20
discovery/azure/azure_test.go
View file

@ -28,6 +28,7 @@ func TestMain(m *testing.M) {
func TestMapFromVMWithEmptyTags(t *testing.T) {
id := "test"
name := "name"
size := "size"
vmType := "type"
location := "westeurope"
computerName := "computer_name"
@ -44,6 +45,9 @@ func TestMapFromVMWithEmptyTags(t *testing.T) {
},
},
NetworkProfile: &networkProfile,
HardwareProfile: &compute.HardwareProfile{
VMSize: compute.VirtualMachineSizeTypes(size),
},
}
testVM := compute.VirtualMachine{
@ -64,6 +68,7 @@ func TestMapFromVMWithEmptyTags(t *testing.T) {
OsType: "Linux",
Tags: map[string]*string{},
NetworkInterfaces: []string{},
Size: size,
}
actualVM := mapFromVM(testVM)
@ -74,6 +79,7 @@ func TestMapFromVMWithEmptyTags(t *testing.T) {
func TestMapFromVMWithTags(t *testing.T) {
id := "test"
name := "name"
size := "size"
vmType := "type"
location := "westeurope"
computerName := "computer_name"
@ -93,6 +99,9 @@ func TestMapFromVMWithTags(t *testing.T) {
},
},
NetworkProfile: &networkProfile,
HardwareProfile: &compute.HardwareProfile{
VMSize: compute.VirtualMachineSizeTypes(size),
},
}
testVM := compute.VirtualMachine{
@ -113,6 +122,7 @@ func TestMapFromVMWithTags(t *testing.T) {
OsType: "Linux",
Tags: tags,
NetworkInterfaces: []string{},
Size: size,
}
actualVM := mapFromVM(testVM)
@ -123,6 +133,7 @@ func TestMapFromVMWithTags(t *testing.T) {
func TestMapFromVMScaleSetVMWithEmptyTags(t *testing.T) {
id := "test"
name := "name"
size := "size"
vmType := "type"
location := "westeurope"
computerName := "computer_name"
@ -139,6 +150,9 @@ func TestMapFromVMScaleSetVMWithEmptyTags(t *testing.T) {
},
},
NetworkProfile: &networkProfile,
HardwareProfile: &compute.HardwareProfile{
VMSize: compute.VirtualMachineSizeTypes(size),
},
}
testVM := compute.VirtualMachineScaleSetVM{
@ -161,6 +175,7 @@ func TestMapFromVMScaleSetVMWithEmptyTags(t *testing.T) {
Tags: map[string]*string{},
NetworkInterfaces: []string{},
ScaleSet: scaleSet,
Size: size,
}
actualVM := mapFromVMScaleSetVM(testVM, scaleSet)
@ -171,6 +186,7 @@ func TestMapFromVMScaleSetVMWithEmptyTags(t *testing.T) {
func TestMapFromVMScaleSetVMWithTags(t *testing.T) {
id := "test"
name := "name"
size := "size"
vmType := "type"
location := "westeurope"
computerName := "computer_name"
@ -190,6 +206,9 @@ func TestMapFromVMScaleSetVMWithTags(t *testing.T) {
},
},
NetworkProfile: &networkProfile,
HardwareProfile: &compute.HardwareProfile{
VMSize: compute.VirtualMachineSizeTypes(size),
},
}
testVM := compute.VirtualMachineScaleSetVM{
@ -212,6 +231,7 @@ func TestMapFromVMScaleSetVMWithTags(t *testing.T) {
Tags: tags,
NetworkInterfaces: []string{},
ScaleSet: scaleSet,
Size: size,
}
actualVM := mapFromVMScaleSetVM(testVM, scaleSet)

9
discovery/consul/consul.go
View file

@ -60,6 +60,8 @@ const (
datacenterLabel = model.MetaLabelPrefix + "consul_dc"
// namespaceLabel is the name of the label containing the namespace (Consul Enterprise only).
namespaceLabel = model.MetaLabelPrefix + "consul_namespace"
// partitionLabel is the name of the label containing the Admin Partition (Consul Enterprise only).
partitionLabel = model.MetaLabelPrefix + "consul_partition"
// taggedAddressesLabel is the prefix for the labels mapping to a target's tagged addresses.
taggedAddressesLabel = model.MetaLabelPrefix + "consul_tagged_address_"
// serviceIDLabel is the name of the label containing the service ID.
@ -109,9 +111,11 @@ func init() {
// SDConfig is the configuration for Consul service discovery.
type SDConfig struct {
Server string `yaml:"server,omitempty"`
PathPrefix string `yaml:"path_prefix,omitempty"`
Token config.Secret `yaml:"token,omitempty"`
Datacenter string `yaml:"datacenter,omitempty"`
Namespace string `yaml:"namespace,omitempty"`
Partition string `yaml:"partition,omitempty"`
TagSeparator string `yaml:"tag_separator,omitempty"`
Scheme string `yaml:"scheme,omitempty"`
Username string `yaml:"username,omitempty"`
@ -183,6 +187,7 @@ type Discovery struct {
client *consul.Client
clientDatacenter string
clientNamespace string
clientPartition string
tagSeparator string
watchedServices []string // Set of services which will be discovered.
watchedTags []string // Tags used to filter instances of a service.
@ -207,9 +212,11 @@ func NewDiscovery(conf *SDConfig, logger log.Logger) (*Discovery, error) {
clientConf := &consul.Config{
Address: conf.Server,
PathPrefix: conf.PathPrefix,
Scheme: conf.Scheme,
Datacenter: conf.Datacenter,
Namespace: conf.Namespace,
Partition: conf.Partition,
Token: string(conf.Token),
HttpClient: wrapper,
}
@ -227,6 +234,7 @@ func NewDiscovery(conf *SDConfig, logger log.Logger) (*Discovery, error) {
refreshInterval: time.Duration(conf.RefreshInterval),
clientDatacenter: conf.Datacenter,
clientNamespace: conf.Namespace,
clientPartition: conf.Partition,
finalizer: wrapper.CloseIdleConnections,
logger: logger,
}
@ -547,6 +555,7 @@ func (srv *consulService) watch(ctx context.Context, ch chan<- []*targetgroup.Gr
addressLabel: model.LabelValue(serviceNode.Node.Address),
nodeLabel: model.LabelValue(serviceNode.Node.Node),
namespaceLabel: model.LabelValue(serviceNode.Service.Namespace),
partitionLabel: model.LabelValue(serviceNode.Service.Partition),
tagsLabel: model.LabelValue(tags),
serviceAddressLabel: model.LabelValue(serviceNode.Service.Address),
servicePortLabel: model.LabelValue(strconv.Itoa(serviceNode.Service.Port)),

2
discovery/consul/consul_test.go
View file

@ -365,7 +365,7 @@ func TestGetDatacenterShouldReturnError(t *testing.T) {
{
// Define a handler that will return status 500.
handler: func(w http.ResponseWriter, r *http.Request) {
w.WriteHeader(500)
w.WriteHeader(http.StatusInternalServerError)
},
errMessage: "Unexpected response code: 500 ()",
},

7
discovery/dns/dns.go
View file

@ -285,21 +285,22 @@ func lookupWithSearchPath(name string, qtype uint16, logger log.Logger) (*dns.Ms
for _, lname := range conf.NameList(name) {
response, err := lookupFromAnyServer(lname, qtype, conf, logger)
if err != nil {
switch {
case err != nil:
// We can't go home yet, because a later name
// may give us a valid, successful answer. However
// we can no longer say "this name definitely doesn't
// exist", because we did not get that answer for
// at least one name.
allResponsesValid = false
} else if response.Rcode == dns.RcodeSuccess {
case response.Rcode == dns.RcodeSuccess:
// Outcome 1: GOLD!
return response, nil
}
}
if allResponsesValid {
// Outcome 2: everyone says NXDOMAIN, that's good enough for me
// Outcome 2: everyone says NXDOMAIN, that's good enough for me.
return &dns.Msg{}, nil
}
// Outcome 3: boned.

20
discovery/file/file.go
View file

@ -39,18 +39,23 @@ import (
)
var (
fileSDReadErrorsCount = prometheus.NewCounter(
prometheus.CounterOpts{
Name: "prometheus_sd_file_read_errors_total",
Help: "The number of File-SD read errors.",
})
fileSDScanDuration = prometheus.NewSummary(
prometheus.SummaryOpts{
Name: "prometheus_sd_file_scan_duration_seconds",
Help: "The duration of the File-SD scan in seconds.",
Objectives: map[float64]float64{0.5: 0.05, 0.9: 0.01, 0.99: 0.001},
})
fileSDReadErrorsCount = prometheus.NewCounter(
fileSDTimeStamp = NewTimestampCollector()
fileWatcherErrorsCount = prometheus.NewCounter(
prometheus.CounterOpts{
Name: "prometheus_sd_file_read_errors_total",
Help: "The number of File-SD read errors.",
Name: "prometheus_sd_file_watcher_errors_total",
Help: "The number of File-SD errors caused by filesystem watch failures.",
})
fileSDTimeStamp = NewTimestampCollector()
patFileSDName = regexp.MustCompile(`^[^*]*(\*[^/]*)?\.(json|yml|yaml|JSON|YML|YAML)$`)
@ -62,7 +67,7 @@ var (
func init() {
discovery.RegisterConfig(&SDConfig{})
prometheus.MustRegister(fileSDScanDuration, fileSDReadErrorsCount, fileSDTimeStamp)
prometheus.MustRegister(fileSDReadErrorsCount, fileSDScanDuration, fileSDTimeStamp, fileWatcherErrorsCount)
}
// SDConfig is the configuration for file based discovery.
@ -221,8 +226,8 @@ func (d *Discovery) watchFiles() {
panic("no watcher configured")
}
for _, p := range d.paths {
if idx := strings.LastIndex(p, "/"); idx > -1 {
p = p[:idx]
if dir, _ := filepath.Split(p); dir != "" {
p = dir
} else {
p = "./"
}
@ -237,6 +242,7 @@ func (d *Discovery) Run(ctx context.Context, ch chan<- []*targetgroup.Group) {
watcher, err := fsnotify.NewWatcher()
if err != nil {
level.Error(d.logger).Log("msg", "Error adding file watcher", "err", err)
fileWatcherErrorsCount.Inc()
return
}
d.watcher = watcher

2
discovery/hetzner/hcloud.go
View file

@ -59,7 +59,7 @@ type hcloudDiscovery struct {
}
// newHcloudDiscovery returns a new hcloudDiscovery which periodically refreshes its targets.
func newHcloudDiscovery(conf *SDConfig, logger log.Logger) (*hcloudDiscovery, error) {
func newHcloudDiscovery(conf *SDConfig, _ log.Logger) (*hcloudDiscovery, error) {
d := &hcloudDiscovery{
port: conf.Port,
}

4
discovery/hetzner/robot.go
View file

@ -51,7 +51,7 @@ type robotDiscovery struct {
}
// newRobotDiscovery returns a new robotDiscovery which periodically refreshes its targets.
func newRobotDiscovery(conf *SDConfig, logger log.Logger) (*robotDiscovery, error) {
func newRobotDiscovery(conf *SDConfig, _ log.Logger) (*robotDiscovery, error) {
d := &robotDiscovery{
port: conf.Port,
endpoint: conf.robotEndpoint,
@ -69,7 +69,7 @@ func newRobotDiscovery(conf *SDConfig, logger log.Logger) (*robotDiscovery, erro
return d, nil
}
func (d *robotDiscovery) refresh(ctx context.Context) ([]*targetgroup.Group, error) {
func (d *robotDiscovery) refresh(context.Context) ([]*targetgroup.Group, error) {
req, err := http.NewRequest("GET", d.endpoint+"/server", nil)
if err != nil {
return nil, err

1
discovery/install/install.go
View file

@ -33,6 +33,7 @@ import (
_ "github.com/prometheus/prometheus/discovery/moby" // register moby
_ "github.com/prometheus/prometheus/discovery/nomad" // register nomad
_ "github.com/prometheus/prometheus/discovery/openstack" // register openstack
_ "github.com/prometheus/prometheus/discovery/ovhcloud" // register ovhcloud
_ "github.com/prometheus/prometheus/discovery/puppetdb" // register puppetdb
_ "github.com/prometheus/prometheus/discovery/scaleway" // register scaleway
_ "github.com/prometheus/prometheus/discovery/triton" // register triton

2
discovery/ionos/server.go
View file

@ -60,7 +60,7 @@ type serverDiscovery struct {
datacenterID string
}
func newServerDiscovery(conf *SDConfig, logger log.Logger) (*serverDiscovery, error) {
func newServerDiscovery(conf *SDConfig, _ log.Logger) (*serverDiscovery, error) {
d := &serverDiscovery{
port: conf.Port,
datacenterID: conf.DatacenterID,

6
discovery/kubernetes/client_metrics.go
View file

@ -122,11 +122,11 @@ func (f *clientGoRequestMetricAdapter) Register(registerer prometheus.Registerer
)
}
func (clientGoRequestMetricAdapter) Increment(ctx context.Context, code, method, host string) {
func (clientGoRequestMetricAdapter) Increment(_ context.Context, code, _, _ string) {
clientGoRequestResultMetricVec.WithLabelValues(code).Inc()
}
func (clientGoRequestMetricAdapter) Observe(ctx context.Context, verb string, u url.URL, latency time.Duration) {
func (clientGoRequestMetricAdapter) Observe(_ context.Context, _ string, u url.URL, latency time.Duration) {
clientGoRequestLatencyMetricVec.WithLabelValues(u.EscapedPath()).Observe(latency.Seconds())
}
@ -169,7 +169,7 @@ func (f *clientGoWorkqueueMetricsProvider) NewLongestRunningProcessorSecondsMetr
return clientGoWorkqueueLongestRunningProcessorMetricVec.WithLabelValues(name)
}
func (clientGoWorkqueueMetricsProvider) NewRetriesMetric(name string) workqueue.CounterMetric {
func (clientGoWorkqueueMetricsProvider) NewRetriesMetric(string) workqueue.CounterMetric {
// Retries are not used so the metric is omitted.
return noopMetric{}
}

46
discovery/kubernetes/endpoints.go
View file

@ -11,6 +11,7 @@
// See the License for the specific language governing permissions and
// limitations under the License.
// nolint:revive // Many legitimately empty blocks in this file.
package kubernetes
import (
@ -72,7 +73,7 @@ func NewEndpoints(l log.Logger, eps cache.SharedIndexInformer, svc, pod, node ca
queue: workqueue.NewNamed("endpoints"),
}
e.endpointsInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
_, err := e.endpointsInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
AddFunc: func(o interface{}) {
epAddCount.Inc()
e.enqueue(o)
@ -86,6 +87,9 @@ func NewEndpoints(l log.Logger, eps cache.SharedIndexInformer, svc, pod, node ca
e.enqueue(o)
},
})
if err != nil {
level.Error(l).Log("msg", "Error adding endpoints event handler.", "err", err)
}
serviceUpdate := func(o interface{}) {
svc, err := convertToService(o)
@ -106,7 +110,7 @@ func NewEndpoints(l log.Logger, eps cache.SharedIndexInformer, svc, pod, node ca
level.Error(e.logger).Log("msg", "retrieving endpoints failed", "err", err)
}
}
e.serviceInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
_, err = e.serviceInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
// TODO(fabxc): potentially remove add and delete event handlers. Those should
// be triggered via the endpoint handlers already.
AddFunc: func(o interface{}) {
@ -122,8 +126,11 @@ func NewEndpoints(l log.Logger, eps cache.SharedIndexInformer, svc, pod, node ca
serviceUpdate(o)
},
})
if err != nil {
level.Error(l).Log("msg", "Error adding services event handler.", "err", err)
}
if e.withNodeMetadata {
e.nodeInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
_, err = e.nodeInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
AddFunc: func(o interface{}) {
node := o.(*apiv1.Node)
e.enqueueNode(node.Name)
@ -137,6 +144,9 @@ func NewEndpoints(l log.Logger, eps cache.SharedIndexInformer, svc, pod, node ca
e.enqueueNode(node.Name)
},
})
if err != nil {
level.Error(l).Log("msg", "Error adding nodes event handler.", "err", err)
}
}
return e
@ -295,7 +305,11 @@ func (e *Endpoints) buildEndpoints(eps *apiv1.Endpoints) *targetgroup.Group {
}
if e.withNodeMetadata {
target = addNodeLabels(target, e.nodeInf, e.logger, addr.NodeName)
if addr.NodeName != nil {
target = addNodeLabels(target, e.nodeInf, e.logger, addr.NodeName)
} else if addr.TargetRef != nil && addr.TargetRef.Kind == "Node" {
target = addNodeLabels(target, e.nodeInf, e.logger, &addr.TargetRef.Name)
}
}
pod := e.resolvePodRef(addr.TargetRef)
@ -375,18 +389,21 @@ func (e *Endpoints) buildEndpoints(eps *apiv1.Endpoints) *targetgroup.Group {
continue
}
a := net.JoinHostPort(pe.pod.Status.PodIP, strconv.FormatUint(uint64(cport.ContainerPort), 10))
ports := strconv.FormatUint(uint64(cport.ContainerPort), 10)
// PodIP can be empty when a pod is starting or has been evicted.
if len(pe.pod.Status.PodIP) != 0 {
a := net.JoinHostPort(pe.pod.Status.PodIP, strconv.FormatUint(uint64(cport.ContainerPort), 10))
ports := strconv.FormatUint(uint64(cport.ContainerPort), 10)
target := model.LabelSet{
model.AddressLabel: lv(a),
podContainerNameLabel: lv(c.Name),
podContainerImageLabel: lv(c.Image),
podContainerPortNameLabel: lv(cport.Name),
podContainerPortNumberLabel: lv(ports),
podContainerPortProtocolLabel: lv(string(cport.Protocol)),
target := model.LabelSet{
model.AddressLabel: lv(a),
podContainerNameLabel: lv(c.Name),
podContainerImageLabel: lv(c.Image),
podContainerPortNameLabel: lv(cport.Name),
podContainerPortNumberLabel: lv(ports),
podContainerPortProtocolLabel: lv(string(cport.Protocol)),
}
tg.Targets = append(tg.Targets, target.Merge(podLabels(pe.pod)))
}
tg.Targets = append(tg.Targets, target.Merge(podLabels(pe.pod)))
}
}
}
@ -456,5 +473,6 @@ func addNodeLabels(tg model.LabelSet, nodeInf cache.SharedInformer, logger log.L
nodeLabelset[model.LabelName(nodeLabelPrefix+ln)] = lv(v)
nodeLabelset[model.LabelName(nodeLabelPresentPrefix+ln)] = presentValue
}
return tg.Merge(nodeLabelset)
}

145
discovery/kubernetes/endpoints_test.go
View file

@ -69,6 +69,24 @@ func makeEndpoints() *v1.Endpoints {
},
},
},
{
Addresses: []v1.EndpointAddress{
{
IP: "6.7.8.9",
TargetRef: &v1.ObjectReference{
Kind: "Node",
Name: "barbaz",
},
},
},
Ports: []v1.EndpointPort{
{
Name: "testport",
Port: 9002,
Protocol: v1.ProtocolTCP,
},
},
},
},
}
}
@ -106,6 +124,14 @@ func TestEndpointsDiscoveryBeforeRun(t *testing.T) {
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "false",
},
{
"__address__": "6.7.8.9:9002",
"__meta_kubernetes_endpoint_address_target_kind": "Node",
"__meta_kubernetes_endpoint_address_target_name": "barbaz",
"__meta_kubernetes_endpoint_port_name": "testport",
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "true",
},
},
Labels: model.LabelSet{
"__meta_kubernetes_namespace": "default",
@ -398,6 +424,14 @@ func TestEndpointsDiscoveryWithService(t *testing.T) {
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "false",
},
{
"__address__": "6.7.8.9:9002",
"__meta_kubernetes_endpoint_address_target_kind": "Node",
"__meta_kubernetes_endpoint_address_target_name": "barbaz",
"__meta_kubernetes_endpoint_port_name": "testport",
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "true",
},
},
Labels: model.LabelSet{
"__meta_kubernetes_namespace": "default",
@ -466,6 +500,14 @@ func TestEndpointsDiscoveryWithServiceUpdate(t *testing.T) {
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "false",
},
{
"__address__": "6.7.8.9:9002",
"__meta_kubernetes_endpoint_address_target_kind": "Node",
"__meta_kubernetes_endpoint_address_target_name": "barbaz",
"__meta_kubernetes_endpoint_port_name": "testport",
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "true",
},
},
Labels: model.LabelSet{
"__meta_kubernetes_namespace": "default",
@ -484,8 +526,10 @@ func TestEndpointsDiscoveryWithServiceUpdate(t *testing.T) {
func TestEndpointsDiscoveryWithNodeMetadata(t *testing.T) {
metadataConfig := AttachMetadataConfig{Node: true}
nodeLabels := map[string]string{"az": "us-east1"}
node := makeNode("foobar", "", "", nodeLabels, nil)
nodeLabels1 := map[string]string{"az": "us-east1"}
nodeLabels2 := map[string]string{"az": "us-west2"}
node1 := makeNode("foobar", "", "", nodeLabels1, nil)
node2 := makeNode("barbaz", "", "", nodeLabels2, nil)
svc := &v1.Service{
ObjectMeta: metav1.ObjectMeta{
Name: "testendpoints",
@ -495,7 +539,7 @@ func TestEndpointsDiscoveryWithNodeMetadata(t *testing.T) {
},
},
}
n, _ := makeDiscoveryWithMetadata(RoleEndpoint, NamespaceDiscovery{}, metadataConfig, makeEndpoints(), svc, node)
n, _ := makeDiscoveryWithMetadata(RoleEndpoint, NamespaceDiscovery{}, metadataConfig, makeEndpoints(), svc, node1, node2)
k8sDiscoveryTest{
discovery: n,
@ -526,6 +570,17 @@ func TestEndpointsDiscoveryWithNodeMetadata(t *testing.T) {
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "false",
},
{
"__address__": "6.7.8.9:9002",
"__meta_kubernetes_endpoint_address_target_kind": "Node",
"__meta_kubernetes_endpoint_address_target_name": "barbaz",
"__meta_kubernetes_endpoint_port_name": "testport",
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "true",
"__meta_kubernetes_node_label_az": "us-west2",
"__meta_kubernetes_node_labelpresent_az": "true",
"__meta_kubernetes_node_name": "barbaz",
},
},
Labels: model.LabelSet{
"__meta_kubernetes_namespace": "default",
@ -541,8 +596,10 @@ func TestEndpointsDiscoveryWithNodeMetadata(t *testing.T) {
}
func TestEndpointsDiscoveryWithUpdatedNodeMetadata(t *testing.T) {
nodeLabels := map[string]string{"az": "us-east1"}
nodes := makeNode("foobar", "", "", nodeLabels, nil)
nodeLabels1 := map[string]string{"az": "us-east1"}
nodeLabels2 := map[string]string{"az": "us-west2"}
node1 := makeNode("foobar", "", "", nodeLabels1, nil)
node2 := makeNode("barbaz", "", "", nodeLabels2, nil)
metadataConfig := AttachMetadataConfig{Node: true}
svc := &v1.Service{
ObjectMeta: metav1.ObjectMeta{
@ -553,13 +610,13 @@ func TestEndpointsDiscoveryWithUpdatedNodeMetadata(t *testing.T) {
},
},
}
n, c := makeDiscoveryWithMetadata(RoleEndpoint, NamespaceDiscovery{}, metadataConfig, makeEndpoints(), nodes, svc)
n, c := makeDiscoveryWithMetadata(RoleEndpoint, NamespaceDiscovery{}, metadataConfig, makeEndpoints(), node1, node2, svc)
k8sDiscoveryTest{
discovery: n,
afterStart: func() {
nodes.Labels["az"] = "eu-central1"
c.CoreV1().Nodes().Update(context.Background(), nodes, metav1.UpdateOptions{})
node1.Labels["az"] = "eu-central1"
c.CoreV1().Nodes().Update(context.Background(), node1, metav1.UpdateOptions{})
},
expectedMaxItems: 2,
expectedRes: map[string]*targetgroup.Group{
@ -572,7 +629,7 @@ func TestEndpointsDiscoveryWithUpdatedNodeMetadata(t *testing.T) {
"__meta_kubernetes_endpoint_port_name": "testport",
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "true",
"__meta_kubernetes_node_label_az": "eu-central1",
"__meta_kubernetes_node_label_az": "us-east1",
"__meta_kubernetes_node_labelpresent_az": "true",
"__meta_kubernetes_node_name": "foobar",
},
@ -588,6 +645,17 @@ func TestEndpointsDiscoveryWithUpdatedNodeMetadata(t *testing.T) {
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "false",
},
{
"__address__": "6.7.8.9:9002",
"__meta_kubernetes_endpoint_address_target_kind": "Node",
"__meta_kubernetes_endpoint_address_target_name": "barbaz",
"__meta_kubernetes_endpoint_port_name": "testport",
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "true",
"__meta_kubernetes_node_label_az": "us-west2",
"__meta_kubernetes_node_labelpresent_az": "true",
"__meta_kubernetes_node_name": "barbaz",
},
},
Labels: model.LabelSet{
"__meta_kubernetes_namespace": "default",
@ -699,6 +767,14 @@ func TestEndpointsDiscoveryNamespaces(t *testing.T) {
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "false",
},
{
"__address__": "6.7.8.9:9002",
"__meta_kubernetes_endpoint_address_target_kind": "Node",
"__meta_kubernetes_endpoint_address_target_name": "barbaz",
"__meta_kubernetes_endpoint_port_name": "testport",
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "true",
},
},
Labels: model.LabelSet{
"__meta_kubernetes_namespace": "ns1",
@ -815,6 +891,14 @@ func TestEndpointsDiscoveryOwnNamespace(t *testing.T) {
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "false",
},
{
"__address__": "6.7.8.9:9002",
"__meta_kubernetes_endpoint_address_target_kind": "Node",
"__meta_kubernetes_endpoint_address_target_name": "barbaz",
"__meta_kubernetes_endpoint_port_name": "testport",
"__meta_kubernetes_endpoint_port_protocol": "TCP",
"__meta_kubernetes_endpoint_ready": "true",
},
},
Labels: model.LabelSet{
"__meta_kubernetes_namespace": "own-ns",
@ -825,3 +909,46 @@ func TestEndpointsDiscoveryOwnNamespace(t *testing.T) {
},
}.Run(t)
}
func TestEndpointsDiscoveryEmptyPodStatus(t *testing.T) {
ep := makeEndpoints()
ep.Namespace = "ns"
pod := &v1.Pod{
ObjectMeta: metav1.ObjectMeta{
Name: "testpod",
Namespace: "ns",
UID: types.UID("deadbeef"),
},
Spec: v1.PodSpec{
NodeName: "testnode",
Containers: []v1.Container{
{
Name: "p1",
Image: "p1:latest",
Ports: []v1.ContainerPort{
{
Name: "mainport",
ContainerPort: 9000,
Protocol: v1.ProtocolTCP,
},
},
},
},
},
Status: v1.PodStatus{},
}
objs := []runtime.Object{
ep,
pod,
}
n, _ := makeDiscovery(RoleEndpoint, NamespaceDiscovery{IncludeOwnNamespace: true}, objs...)
k8sDiscoveryTest{
discovery: n,
expectedMaxItems: 0,
expectedRes: map[string]*targetgroup.Group{},
}.Run(t)
}

58
discovery/kubernetes/endpointslice.go
View file

@ -73,7 +73,7 @@ func NewEndpointSlice(l log.Logger, eps cache.SharedIndexInformer, svc, pod, nod
queue: workqueue.NewNamed("endpointSlice"),
}
e.endpointSliceInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
_, err := e.endpointSliceInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
AddFunc: func(o interface{}) {
epslAddCount.Inc()
e.enqueue(o)
@ -87,6 +87,9 @@ func NewEndpointSlice(l log.Logger, eps cache.SharedIndexInformer, svc, pod, nod
e.enqueue(o)
},
})
if err != nil {
level.Error(l).Log("msg", "Error adding endpoint slices event handler.", "err", err)
}
serviceUpdate := func(o interface{}) {
svc, err := convertToService(o)
@ -109,7 +112,7 @@ func NewEndpointSlice(l log.Logger, eps cache.SharedIndexInformer, svc, pod, nod
}
}
}
e.serviceInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
_, err = e.serviceInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
AddFunc: func(o interface{}) {
svcAddCount.Inc()
serviceUpdate(o)
@ -123,9 +126,12 @@ func NewEndpointSlice(l log.Logger, eps cache.SharedIndexInformer, svc, pod, nod
serviceUpdate(o)
},
})
if err != nil {
level.Error(l).Log("msg", "Error adding services event handler.", "err", err)
}
if e.withNodeMetadata {
e.nodeInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
_, err = e.nodeInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
AddFunc: func(o interface{}) {
node := o.(*apiv1.Node)
e.enqueueNode(node.Name)
@ -139,6 +145,9 @@ func NewEndpointSlice(l log.Logger, eps cache.SharedIndexInformer, svc, pod, nod
e.enqueueNode(node.Name)
},
})
if err != nil {
level.Error(l).Log("msg", "Error adding nodes event handler.", "err", err)
}
}
return e
@ -181,7 +190,7 @@ func (e *EndpointSlice) Run(ctx context.Context, ch chan<- []*targetgroup.Group)
}
go func() {
for e.process(ctx, ch) {
for e.process(ctx, ch) { // nolint:revive
}
}()
@ -250,6 +259,8 @@ const (
endpointSlicePortLabel = metaLabelPrefix + "endpointslice_port"
endpointSlicePortAppProtocol = metaLabelPrefix + "endpointslice_port_app_protocol"
endpointSliceEndpointConditionsReadyLabel = metaLabelPrefix + "endpointslice_endpoint_conditions_ready"
endpointSliceEndpointConditionsServingLabel = metaLabelPrefix + "endpointslice_endpoint_conditions_serving"
endpointSliceEndpointConditionsTerminatingLabel = metaLabelPrefix + "endpointslice_endpoint_conditions_terminating"
endpointSliceEndpointHostnameLabel = metaLabelPrefix + "endpointslice_endpoint_hostname"
endpointSliceAddressTargetKindLabel = metaLabelPrefix + "endpointslice_address_target_kind"
endpointSliceAddressTargetNameLabel = metaLabelPrefix + "endpointslice_address_target_name"
@ -289,7 +300,7 @@ func (e *EndpointSlice) buildEndpointSlice(eps endpointSliceAdaptor) *targetgrou
}
if port.protocol() != nil {
target[endpointSlicePortProtocolLabel] = lv(string(*port.protocol()))
target[endpointSlicePortProtocolLabel] = lv(*port.protocol())
}
if port.port() != nil {
@ -304,6 +315,14 @@ func (e *EndpointSlice) buildEndpointSlice(eps endpointSliceAdaptor) *targetgrou
target[endpointSliceEndpointConditionsReadyLabel] = lv(strconv.FormatBool(*ep.conditions().ready()))
}
if ep.conditions().serving() != nil {
target[endpointSliceEndpointConditionsServingLabel] = lv(strconv.FormatBool(*ep.conditions().serving()))
}
if ep.conditions().terminating() != nil {
target[endpointSliceEndpointConditionsTerminatingLabel] = lv(strconv.FormatBool(*ep.conditions().terminating()))
}
if ep.hostname() != nil {
target[endpointSliceEndpointHostnameLabel] = lv(*ep.hostname())
}
@ -320,7 +339,11 @@ func (e *EndpointSlice) buildEndpointSlice(eps endpointSliceAdaptor) *targetgrou
}
if e.withNodeMetadata {
target = addNodeLabels(target, e.nodeInf, e.logger, ep.nodename())
if ep.targetRef() != nil && ep.targetRef().Kind == "Node" {
target = addNodeLabels(target, e.nodeInf, e.logger, &ep.targetRef().Name)
} else {
target = addNodeLabels(target, e.nodeInf, e.logger, ep.nodename())
}
}
pod := e.resolvePodRef(ep.targetRef())
@ -393,18 +416,21 @@ func (e *EndpointSlice) buildEndpointSlice(eps endpointSliceAdaptor) *targetgrou
continue
}
a := net.JoinHostPort(pe.pod.Status.PodIP, strconv.FormatUint(uint64(cport.ContainerPort), 10))
ports := strconv.FormatUint(uint64(cport.ContainerPort), 10)
// PodIP can be empty when a pod is starting or has been evicted.
if len(pe.pod.Status.PodIP) != 0 {
a := net.JoinHostPort(pe.pod.Status.PodIP, strconv.FormatUint(uint64(cport.ContainerPort), 10))
ports := strconv.FormatUint(uint64(cport.ContainerPort), 10)
target := model.LabelSet{
model.AddressLabel: lv(a),
podContainerNameLabel: lv(c.Name),
podContainerImageLabel: lv(c.Image),
podContainerPortNameLabel: lv(cport.Name),
podContainerPortNumberLabel: lv(ports),
podContainerPortProtocolLabel: lv(string(cport.Protocol)),
target := model.LabelSet{
model.AddressLabel: lv(a),
podContainerNameLabel: lv(c.Name),
podContainerImageLabel: lv(c.Image),
podContainerPortNameLabel: lv(cport.Name),
podContainerPortNumberLabel: lv(ports),
podContainerPortProtocolLabel: lv(string(cport.Protocol)),
}
tg.Targets = append(tg.Targets, target.Merge(podLabels(pe.pod)))
}
tg.Targets = append(tg.Targets, target.Merge(podLabels(pe.pod)))
}
}
}

18
discovery/kubernetes/endpointslice_adaptor.go
View file

@ -49,6 +49,8 @@ type endpointSliceEndpointAdaptor interface {
type endpointSliceEndpointConditionsAdaptor interface {
ready() *bool
serving() *bool
terminating() *bool
}
// Adaptor for k8s.io/api/discovery/v1
@ -193,6 +195,14 @@ func (e *endpointSliceEndpointConditionsAdaptorV1) ready() *bool {
return e.endpointConditions.Ready
}
func (e *endpointSliceEndpointConditionsAdaptorV1) serving() *bool {
return e.endpointConditions.Serving
}
func (e *endpointSliceEndpointConditionsAdaptorV1) terminating() *bool {
return e.endpointConditions.Terminating
}
type endpointSliceEndpointAdaptorV1beta1 struct {
endpoint v1beta1.Endpoint
}
@ -237,6 +247,14 @@ func (e *endpointSliceEndpointConditionsAdaptorV1beta1) ready() *bool {
return e.endpointConditions.Ready
}
func (e *endpointSliceEndpointConditionsAdaptorV1beta1) serving() *bool {
return e.endpointConditions.Serving
}
func (e *endpointSliceEndpointConditionsAdaptorV1beta1) terminating() *bool {
return e.endpointConditions.Terminating
}
type endpointSlicePortAdaptorV1 struct {
endpointPort v1.EndpointPort
}

4
discovery/kubernetes/endpointslice_adaptor_test.go
View file

@ -35,6 +35,8 @@ func Test_EndpointSliceAdaptor_v1(t *testing.T) {
require.Equal(t, endpointSlice.Endpoints[i].Addresses, endpointAdaptor.addresses())
require.Equal(t, endpointSlice.Endpoints[i].Hostname, endpointAdaptor.hostname())
require.Equal(t, endpointSlice.Endpoints[i].Conditions.Ready, endpointAdaptor.conditions().ready())
require.Equal(t, endpointSlice.Endpoints[i].Conditions.Serving, endpointAdaptor.conditions().serving())
require.Equal(t, endpointSlice.Endpoints[i].Conditions.Terminating, endpointAdaptor.conditions().terminating())
require.Equal(t, endpointSlice.Endpoints[i].TargetRef, endpointAdaptor.targetRef())
require.Equal(t, endpointSlice.Endpoints[i].DeprecatedTopology, endpointAdaptor.topology())
}
@ -61,6 +63,8 @@ func Test_EndpointSliceAdaptor_v1beta1(t *testing.T) {
require.Equal(t, endpointSlice.Endpoints[i].Addresses, endpointAdaptor.addresses())
require.Equal(t, endpointSlice.Endpoints[i].Hostname, endpointAdaptor.hostname())
require.Equal(t, endpointSlice.Endpoints[i].Conditions.Ready, endpointAdaptor.conditions().ready())
require.Equal(t, endpointSlice.Endpoints[i].Conditions.Serving, endpointAdaptor.conditions().serving())
require.Equal(t, endpointSlice.Endpoints[i].Conditions.Terminating, endpointAdaptor.conditions().terminating())
require.Equal(t, endpointSlice.Endpoints[i].TargetRef, endpointAdaptor.targetRef())
require.Equal(t, endpointSlice.Endpoints[i].Topology, endpointAdaptor.topology())
}

531
discovery/kubernetes/endpointslice_test.go
View file

@ -64,23 +64,42 @@ func makeEndpointSliceV1() *v1.EndpointSlice {
},
Endpoints: []v1.Endpoint{
{
Addresses: []string{"1.2.3.4"},
Conditions: v1.EndpointConditions{Ready: boolptr(true)},
Hostname: strptr("testendpoint1"),
TargetRef: &corev1.ObjectReference{},
NodeName: strptr("foobar"),
Addresses: []string{"1.2.3.4"},
Conditions: v1.EndpointConditions{
Ready: boolptr(true),
Serving: boolptr(true),
Terminating: boolptr(false),
},
Hostname: strptr("testendpoint1"),
TargetRef: &corev1.ObjectReference{},
NodeName: strptr("foobar"),
DeprecatedTopology: map[string]string{
"topology": "value",
},
}, {
Addresses: []string{"2.3.4.5"},
Conditions: v1.EndpointConditions{
Ready: boolptr(true),
Ready: boolptr(true),
Serving: boolptr(true),
Terminating: boolptr(false),
},
}, {
Addresses: []string{"3.4.5.6"},
Conditions: v1.EndpointConditions{
Ready: boolptr(false),
Ready: boolptr(false),
Serving: boolptr(true),
Terminating: boolptr(true),
},
}, {
Addresses: []string{"4.5.6.7"},
Conditions: v1.EndpointConditions{
Ready: boolptr(true),
Serving: boolptr(true),
Terminating: boolptr(false),
},
TargetRef: &corev1.ObjectReference{
Kind: "Node",
Name: "barbaz",
},
},
},
@ -111,12 +130,27 @@ func makeEndpointSliceV1beta1() *v1beta1.EndpointSlice {
}, {
Addresses: []string{"2.3.4.5"},
Conditions: v1beta1.EndpointConditions{
Ready: boolptr(true),
Ready: boolptr(true),
Serving: boolptr(true),
Terminating: boolptr(false),
},
}, {
Addresses: []string{"3.4.5.6"},
Conditions: v1beta1.EndpointConditions{
Ready: boolptr(false),
Ready: boolptr(false),
Serving: boolptr(true),
Terminating: boolptr(true),
},
}, {
Addresses: []string{"4.5.6.7"},
Conditions: v1beta1.EndpointConditions{
Ready: boolptr(true),
Serving: boolptr(true),
Terminating: boolptr(false),
},
TargetRef: &corev1.ObjectReference{
Kind: "Node",
Name: "barbaz",
},
},
},
@ -141,6 +175,8 @@ func TestEndpointSliceDiscoveryBeforeRun(t *testing.T) {
"__meta_kubernetes_endpointslice_address_target_kind": "",
"__meta_kubernetes_endpointslice_address_target_name": "",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_endpoint_hostname": "testendpoint1",
"__meta_kubernetes_endpointslice_endpoint_topology_present_topology": "true",
"__meta_kubernetes_endpointslice_endpoint_topology_topology": "value",
@ -151,19 +187,35 @@ func TestEndpointSliceDiscoveryBeforeRun(t *testing.T) {
},
{
"__address__": "2.3.4.5:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "3.4.5.6:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "4.5.6.7:9000",
"__meta_kubernetes_endpointslice_address_target_kind": "Node",
"__meta_kubernetes_endpointslice_address_target_name": "barbaz",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
},
Labels: model.LabelSet{
@ -199,17 +251,32 @@ func TestEndpointSliceDiscoveryBeforeRunV1beta1(t *testing.T) {
},
{
"__address__": "2.3.4.5:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "3.4.5.6:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "4.5.6.7:9000",
"__meta_kubernetes_endpointslice_address_target_kind": "Node",
"__meta_kubernetes_endpointslice_address_target_name": "barbaz",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
},
Labels: model.LabelSet{
@ -367,6 +434,8 @@ func TestEndpointSliceDiscoveryDelete(t *testing.T) {
"__meta_kubernetes_endpointslice_address_target_kind": "",
"__meta_kubernetes_endpointslice_address_target_name": "",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_endpoint_hostname": "testendpoint1",
"__meta_kubernetes_endpointslice_endpoint_topology_present_topology": "true",
"__meta_kubernetes_endpointslice_endpoint_topology_topology": "value",
@ -377,19 +446,35 @@ func TestEndpointSliceDiscoveryDelete(t *testing.T) {
},
{
"__address__": "2.3.4.5:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "3.4.5.6:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "4.5.6.7:9000",
"__meta_kubernetes_endpointslice_address_target_kind": "Node",
"__meta_kubernetes_endpointslice_address_target_name": "barbaz",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
},
Labels: map[model.LabelName]model.LabelValue{
@ -445,6 +530,8 @@ func TestEndpointSliceDiscoveryUpdate(t *testing.T) {
"__meta_kubernetes_endpointslice_address_target_kind": "",
"__meta_kubernetes_endpointslice_address_target_name": "",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_endpoint_hostname": "testendpoint1",
"__meta_kubernetes_endpointslice_endpoint_topology_present_topology": "true",
"__meta_kubernetes_endpointslice_endpoint_topology_topology": "value",
@ -455,19 +542,35 @@ func TestEndpointSliceDiscoveryUpdate(t *testing.T) {
},
{
"__address__": "2.3.4.5:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "3.4.5.6:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "4.5.6.7:9000",
"__meta_kubernetes_endpointslice_address_target_kind": "Node",
"__meta_kubernetes_endpointslice_address_target_name": "barbaz",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
},
Labels: model.LabelSet{
@ -512,6 +615,8 @@ func TestEndpointSliceDiscoveryEmptyEndpoints(t *testing.T) {
"__meta_kubernetes_endpointslice_address_target_kind": "",
"__meta_kubernetes_endpointslice_address_target_name": "",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_endpoint_hostname": "testendpoint1",
"__meta_kubernetes_endpointslice_endpoint_topology_present_topology": "true",
"__meta_kubernetes_endpointslice_endpoint_topology_topology": "value",
@ -522,19 +627,35 @@ func TestEndpointSliceDiscoveryEmptyEndpoints(t *testing.T) {
},
{
"__address__": "2.3.4.5:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "3.4.5.6:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "4.5.6.7:9000",
"__meta_kubernetes_endpointslice_address_target_kind": "Node",
"__meta_kubernetes_endpointslice_address_target_name": "barbaz",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
},
Labels: model.LabelSet{
@ -574,6 +695,8 @@ func TestEndpointSliceDiscoveryWithService(t *testing.T) {
"__meta_kubernetes_endpointslice_address_target_kind": "",
"__meta_kubernetes_endpointslice_address_target_name": "",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_endpoint_hostname": "testendpoint1",
"__meta_kubernetes_endpointslice_endpoint_topology_present_topology": "true",
"__meta_kubernetes_endpointslice_endpoint_topology_topology": "value",
@ -584,19 +707,35 @@ func TestEndpointSliceDiscoveryWithService(t *testing.T) {
},
{
"__address__": "2.3.4.5:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "3.4.5.6:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "4.5.6.7:9000",
"__meta_kubernetes_endpointslice_address_target_kind": "Node",
"__meta_kubernetes_endpointslice_address_target_name": "barbaz",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
},
Labels: model.LabelSet{
@ -652,6 +791,8 @@ func TestEndpointSliceDiscoveryWithServiceUpdate(t *testing.T) {
"__meta_kubernetes_endpointslice_address_target_kind": "",
"__meta_kubernetes_endpointslice_address_target_name": "",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_endpoint_hostname": "testendpoint1",
"__meta_kubernetes_endpointslice_endpoint_topology_present_topology": "true",
"__meta_kubernetes_endpointslice_endpoint_topology_topology": "value",
@ -662,19 +803,35 @@ func TestEndpointSliceDiscoveryWithServiceUpdate(t *testing.T) {
},
{
"__address__": "2.3.4.5:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
},
{
"__address__": "3.4.5.6:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
},
{
"__address__": "4.5.6.7:9000",
"__meta_kubernetes_endpointslice_address_target_kind": "Node",
"__meta_kubernetes_endpointslice_address_target_name": "barbaz",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
},
Labels: model.LabelSet{
@ -695,7 +852,8 @@ func TestEndpointSliceDiscoveryWithServiceUpdate(t *testing.T) {
func TestEndpointsSlicesDiscoveryWithNodeMetadata(t *testing.T) {
metadataConfig := AttachMetadataConfig{Node: true}
nodeLabels := map[string]string{"az": "us-east1"}
nodeLabels1 := map[string]string{"az": "us-east1"}
nodeLabels2 := map[string]string{"az": "us-west2"}
svc := &corev1.Service{
ObjectMeta: metav1.ObjectMeta{
Name: "testendpoints",
@ -705,7 +863,7 @@ func TestEndpointsSlicesDiscoveryWithNodeMetadata(t *testing.T) {
},
},
}
objs := []runtime.Object{makeEndpointSliceV1(), makeNode("foobar", "", "", nodeLabels, nil), svc}
objs := []runtime.Object{makeEndpointSliceV1(), makeNode("foobar", "", "", nodeLabels1, nil), makeNode("barbaz", "", "", nodeLabels2, nil), svc}
n, _ := makeDiscoveryWithMetadata(RoleEndpointSlice, NamespaceDiscovery{}, metadataConfig, objs...)
k8sDiscoveryTest{
@ -719,6 +877,8 @@ func TestEndpointsSlicesDiscoveryWithNodeMetadata(t *testing.T) {
"__meta_kubernetes_endpointslice_address_target_kind": "",
"__meta_kubernetes_endpointslice_address_target_name": "",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_endpoint_hostname": "testendpoint1",
"__meta_kubernetes_endpointslice_endpoint_topology_present_topology": "true",
"__meta_kubernetes_endpointslice_endpoint_topology_topology": "value",
@ -732,19 +892,38 @@ func TestEndpointsSlicesDiscoveryWithNodeMetadata(t *testing.T) {
},
{
"__address__": "2.3.4.5:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "3.4.5.6:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "4.5.6.7:9000",
"__meta_kubernetes_endpointslice_address_target_kind": "Node",
"__meta_kubernetes_endpointslice_address_target_name": "barbaz",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_node_label_az": "us-west2",
"__meta_kubernetes_node_labelpresent_az": "true",
"__meta_kubernetes_node_name": "barbaz",
},
},
Labels: model.LabelSet{
@ -763,7 +942,8 @@ func TestEndpointsSlicesDiscoveryWithNodeMetadata(t *testing.T) {
func TestEndpointsSlicesDiscoveryWithUpdatedNodeMetadata(t *testing.T) {
metadataConfig := AttachMetadataConfig{Node: true}
nodeLabels := map[string]string{"az": "us-east1"}
nodeLabels1 := map[string]string{"az": "us-east1"}
nodeLabels2 := map[string]string{"az": "us-west2"}
svc := &corev1.Service{
ObjectMeta: metav1.ObjectMeta{
Name: "testendpoints",
@ -773,16 +953,17 @@ func TestEndpointsSlicesDiscoveryWithUpdatedNodeMetadata(t *testing.T) {
},
},
}
node := makeNode("foobar", "", "", nodeLabels, nil)
objs := []runtime.Object{makeEndpointSliceV1(), node, svc}
node1 := makeNode("foobar", "", "", nodeLabels1, nil)
node2 := makeNode("barbaz", "", "", nodeLabels2, nil)
objs := []runtime.Object{makeEndpointSliceV1(), node1, node2, svc}
n, c := makeDiscoveryWithMetadata(RoleEndpointSlice, NamespaceDiscovery{}, metadataConfig, objs...)
k8sDiscoveryTest{
discovery: n,
expectedMaxItems: 2,
afterStart: func() {
node.Labels["az"] = "us-central1"
c.CoreV1().Nodes().Update(context.Background(), node, metav1.UpdateOptions{})
node1.Labels["az"] = "us-central1"
c.CoreV1().Nodes().Update(context.Background(), node1, metav1.UpdateOptions{})
},
expectedRes: map[string]*targetgroup.Group{
"endpointslice/default/testendpoints": {
@ -792,6 +973,8 @@ func TestEndpointsSlicesDiscoveryWithUpdatedNodeMetadata(t *testing.T) {
"__meta_kubernetes_endpointslice_address_target_kind": "",
"__meta_kubernetes_endpointslice_address_target_name": "",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_endpoint_hostname": "testendpoint1",
"__meta_kubernetes_endpointslice_endpoint_topology_present_topology": "true",
"__meta_kubernetes_endpointslice_endpoint_topology_topology": "value",
@ -799,25 +982,44 @@ func TestEndpointsSlicesDiscoveryWithUpdatedNodeMetadata(t *testing.T) {
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_node_label_az": "us-central1",
"__meta_kubernetes_node_label_az": "us-east1",
"__meta_kubernetes_node_labelpresent_az": "true",
"__meta_kubernetes_node_name": "foobar",
},
{
"__address__": "2.3.4.5:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "3.4.5.6:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
{
"__address__": "4.5.6.7:9000",
"__meta_kubernetes_endpointslice_address_target_kind": "Node",
"__meta_kubernetes_endpointslice_address_target_name": "barbaz",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_node_label_az": "us-west2",
"__meta_kubernetes_node_labelpresent_az": "true",
"__meta_kubernetes_node_name": "barbaz",
},
},
Labels: model.LabelSet{
@ -913,6 +1115,8 @@ func TestEndpointSliceDiscoveryNamespaces(t *testing.T) {
"__meta_kubernetes_endpointslice_address_target_kind": "",
"__meta_kubernetes_endpointslice_address_target_name": "",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_endpoint_hostname": "testendpoint1",
"__meta_kubernetes_endpointslice_endpoint_topology_present_topology": "true",
"__meta_kubernetes_endpointslice_endpoint_topology_topology": "value",
@ -923,19 +1127,35 @@ func TestEndpointSliceDiscoveryNamespaces(t *testing.T) {
},
{
"__address__": "2.3.4.5:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
},
{
"__address__": "3.4.5.6:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
},
{
"__address__": "4.5.6.7:9000",
"__meta_kubernetes_endpointslice_address_target_kind": "Node",
"__meta_kubernetes_endpointslice_address_target_name": "barbaz",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
},
Labels: model.LabelSet{
@ -1039,6 +1259,8 @@ func TestEndpointSliceDiscoveryOwnNamespace(t *testing.T) {
"__meta_kubernetes_endpointslice_address_target_kind": "",
"__meta_kubernetes_endpointslice_address_target_name": "",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_endpoint_hostname": "testendpoint1",
"__meta_kubernetes_endpointslice_endpoint_topology_present_topology": "true",
"__meta_kubernetes_endpointslice_endpoint_topology_topology": "value",
@ -1049,19 +1271,35 @@ func TestEndpointSliceDiscoveryOwnNamespace(t *testing.T) {
},
{
"__address__": "2.3.4.5:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
},
{
"__address__": "3.4.5.6:9000",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "false",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "true",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
},
{
"__address__": "4.5.6.7:9000",
"__meta_kubernetes_endpointslice_address_target_kind": "Node",
"__meta_kubernetes_endpointslice_address_target_name": "barbaz",
"__meta_kubernetes_endpointslice_endpoint_conditions_ready": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_serving": "true",
"__meta_kubernetes_endpointslice_endpoint_conditions_terminating": "false",
"__meta_kubernetes_endpointslice_port": "9000",
"__meta_kubernetes_endpointslice_port_app_protocol": "http",
"__meta_kubernetes_endpointslice_port_name": "testport",
"__meta_kubernetes_endpointslice_port_protocol": "TCP",
},
},
Labels: model.LabelSet{
@ -1074,3 +1312,46 @@ func TestEndpointSliceDiscoveryOwnNamespace(t *testing.T) {
},
}.Run(t)
}
func TestEndpointSliceDiscoveryEmptyPodStatus(t *testing.T) {
ep := makeEndpointSliceV1()
ep.Namespace = "ns"
pod := &corev1.Pod{
ObjectMeta: metav1.ObjectMeta{
Name: "testpod",
Namespace: "ns",
UID: types.UID("deadbeef"),
},
Spec: corev1.PodSpec{
NodeName: "testnode",
Containers: []corev1.Container{
{
Name: "p1",
Image: "p1:latest",
Ports: []corev1.ContainerPort{
{
Name: "mainport",
ContainerPort: 9000,
Protocol: corev1.ProtocolTCP,
},
},
},
},
},
Status: corev1.PodStatus{},
}
objs := []runtime.Object{
ep,
pod,
}
n, _ := makeDiscovery(RoleEndpoint, NamespaceDiscovery{IncludeOwnNamespace: true}, objs...)
k8sDiscoveryTest{
discovery: n,
expectedMaxItems: 0,
expectedRes: map[string]*targetgroup.Group{},
}.Run(t)
}

7
discovery/kubernetes/ingress.go
View file

@ -48,7 +48,7 @@ type Ingress struct {
// NewIngress returns a new ingress discovery.
func NewIngress(l log.Logger, inf cache.SharedInformer) *Ingress {
s := &Ingress{logger: l, informer: inf, store: inf.GetStore(), queue: workqueue.NewNamed("ingress")}
s.informer.AddEventHandler(cache.ResourceEventHandlerFuncs{
_, err := s.informer.AddEventHandler(cache.ResourceEventHandlerFuncs{
AddFunc: func(o interface{}) {
ingressAddCount.Inc()
s.enqueue(o)
@ -62,6 +62,9 @@ func NewIngress(l log.Logger, inf cache.SharedInformer) *Ingress {
s.enqueue(o)
},
})
if err != nil {
level.Error(l).Log("msg", "Error adding ingresses event handler.", "err", err)
}
return s
}
@ -86,7 +89,7 @@ func (i *Ingress) Run(ctx context.Context, ch chan<- []*targetgroup.Group) {
}
go func() {
for i.process(ctx, ch) {
for i.process(ctx, ch) { // nolint:revive
}
}()

74
discovery/kubernetes/kubernetes.go
View file

@ -299,12 +299,13 @@ func New(l log.Logger, conf *SDConfig) (*Discovery, error) {
err error
ownNamespace string
)
if conf.KubeConfig != "" {
switch {
case conf.KubeConfig != "":
kcfg, err = clientcmd.BuildConfigFromFlags("", conf.KubeConfig)
if err != nil {
return nil, err
}
} else if conf.APIServer.URL == nil {
case conf.APIServer.URL == nil:
// Use the Kubernetes provided pod service account
// as described in https://kubernetes.io/docs/admin/service-accounts-admin/
kcfg, err = rest.InClusterConfig()
@ -324,7 +325,7 @@ func New(l log.Logger, conf *SDConfig) (*Discovery, error) {
}
level.Info(l).Log("msg", "Using pod service account via in-cluster config")
} else {
default:
rt, err := config.NewRoundTripperFromConfig(conf.HTTPClientConfig, "kubernetes_sd")
if err != nil {
return nil, err
@ -382,7 +383,8 @@ func mapSelector(rawSelector []SelectorConfig) roleSelector {
return rs
}
const resyncPeriod = 10 * time.Minute
// Disable the informer's resync, which just periodically resends already processed updates and distort SD metrics.
const resyncDisabled = 0
// Run implements the discoverer interface.
func (d *Discovery) Run(ctx context.Context, ch chan<- []*targetgroup.Group) {
@ -475,8 +477,8 @@ func (d *Discovery) Run(ctx context.Context, ch chan<- []*targetgroup.Group) {
eps := NewEndpointSlice(
log.With(d.logger, "role", "endpointslice"),
informer,
cache.NewSharedInformer(slw, &apiv1.Service{}, resyncPeriod),
cache.NewSharedInformer(plw, &apiv1.Pod{}, resyncPeriod),
cache.NewSharedInformer(slw, &apiv1.Service{}, resyncDisabled),
cache.NewSharedInformer(plw, &apiv1.Pod{}, resyncDisabled),
nodeInf,
)
d.discoverers = append(d.discoverers, eps)
@ -534,8 +536,8 @@ func (d *Discovery) Run(ctx context.Context, ch chan<- []*targetgroup.Group) {
eps := NewEndpoints(
log.With(d.logger, "role", "endpoint"),
d.newEndpointsByNodeInformer(elw),
cache.NewSharedInformer(slw, &apiv1.Service{}, resyncPeriod),
cache.NewSharedInformer(plw, &apiv1.Pod{}, resyncPeriod),
cache.NewSharedInformer(slw, &apiv1.Service{}, resyncDisabled),
cache.NewSharedInformer(plw, &apiv1.Pod{}, resyncDisabled),
nodeInf,
)
d.discoverers = append(d.discoverers, eps)
@ -589,7 +591,7 @@ func (d *Discovery) Run(ctx context.Context, ch chan<- []*targetgroup.Group) {
}
svc := NewService(
log.With(d.logger, "role", "service"),
cache.NewSharedInformer(slw, &apiv1.Service{}, resyncPeriod),
cache.NewSharedInformer(slw, &apiv1.Service{}, resyncDisabled),
)
d.discoverers = append(d.discoverers, svc)
go svc.informer.Run(ctx.Done())
@ -627,7 +629,7 @@ func (d *Discovery) Run(ctx context.Context, ch chan<- []*targetgroup.Group) {
return i.Watch(ctx, options)
},
}
informer = cache.NewSharedInformer(ilw, &networkv1.Ingress{}, resyncPeriod)
informer = cache.NewSharedInformer(ilw, &networkv1.Ingress{}, resyncDisabled)
} else {
i := d.client.NetworkingV1beta1().Ingresses(namespace)
ilw := &cache.ListWatch{
@ -642,7 +644,7 @@ func (d *Discovery) Run(ctx context.Context, ch chan<- []*targetgroup.Group) {
return i.Watch(ctx, options)
},
}
informer = cache.NewSharedInformer(ilw, &v1beta1.Ingress{}, resyncPeriod)
informer = cache.NewSharedInformer(ilw, &v1beta1.Ingress{}, resyncDisabled)
}
ingress := NewIngress(
log.With(d.logger, "role", "ingress"),
@ -732,7 +734,7 @@ func (d *Discovery) newNodeInformer(ctx context.Context) cache.SharedInformer {
return d.client.CoreV1().Nodes().Watch(ctx, options)
},
}
return cache.NewSharedInformer(nlw, &apiv1.Node{}, resyncPeriod)
return cache.NewSharedInformer(nlw, &apiv1.Node{}, resyncDisabled)
}
func (d *Discovery) newPodsByNodeInformer(plw *cache.ListWatch) cache.SharedIndexInformer {
@ -747,39 +749,45 @@ func (d *Discovery) newPodsByNodeInformer(plw *cache.ListWatch) cache.SharedInde
}
}
return cache.NewSharedIndexInformer(plw, &apiv1.Pod{}, resyncPeriod, indexers)
return cache.NewSharedIndexInformer(plw, &apiv1.Pod{}, resyncDisabled, indexers)
}
func (d *Discovery) newEndpointsByNodeInformer(plw *cache.ListWatch) cache.SharedIndexInformer {
indexers := make(map[string]cache.IndexFunc)
if !d.attachMetadata.Node {
return cache.NewSharedIndexInformer(plw, &apiv1.Endpoints{}, resyncPeriod, indexers)
return cache.NewSharedIndexInformer(plw, &apiv1.Endpoints{}, resyncDisabled, indexers)
}
indexers[nodeIndex] = func(obj interface{}) ([]string, error) {
e, ok := obj.(*apiv1.Endpoints)
if !ok {
return nil, fmt.Errorf("object is not a pod")
return nil, fmt.Errorf("object is not endpoints")
}
var nodes []string
for _, target := range e.Subsets {
for _, addr := range target.Addresses {
if addr.NodeName == nil {
continue
if addr.TargetRef != nil {
switch addr.TargetRef.Kind {
case "Pod":
if addr.NodeName != nil {
nodes = append(nodes, *addr.NodeName)
}
case "Node":
nodes = append(nodes, addr.TargetRef.Name)
}
}
nodes = append(nodes, *addr.NodeName)
}
}
return nodes, nil
}
return cache.NewSharedIndexInformer(plw, &apiv1.Endpoints{}, resyncPeriod, indexers)
return cache.NewSharedIndexInformer(plw, &apiv1.Endpoints{}, resyncDisabled, indexers)
}
func (d *Discovery) newEndpointSlicesByNodeInformer(plw *cache.ListWatch, object runtime.Object) cache.SharedIndexInformer {
indexers := make(map[string]cache.IndexFunc)
if !d.attachMetadata.Node {
cache.NewSharedIndexInformer(plw, &disv1.EndpointSlice{}, resyncPeriod, indexers)
cache.NewSharedIndexInformer(plw, &disv1.EndpointSlice{}, resyncDisabled, indexers)
}
indexers[nodeIndex] = func(obj interface{}) ([]string, error) {
@ -787,17 +795,29 @@ func (d *Discovery) newEndpointSlicesByNodeInformer(plw *cache.ListWatch, object
switch e := obj.(type) {
case *disv1.EndpointSlice:
for _, target := range e.Endpoints {
if target.NodeName == nil {
continue
if target.TargetRef != nil {
switch target.TargetRef.Kind {
case "Pod":
if target.NodeName != nil {
nodes = append(nodes, *target.NodeName)
}
case "Node":
nodes = append(nodes, target.TargetRef.Name)
}
}
nodes = append(nodes, *target.NodeName)
}
case *disv1beta1.EndpointSlice:
for _, target := range e.Endpoints {
if target.NodeName == nil {
continue
if target.TargetRef != nil {
switch target.TargetRef.Kind {
case "Pod":
if target.NodeName != nil {
nodes = append(nodes, *target.NodeName)
}
case "Node":
nodes = append(nodes, target.TargetRef.Name)
}
}
nodes = append(nodes, *target.NodeName)
}
default:
return nil, fmt.Errorf("object is not an endpointslice")
@ -806,7 +826,7 @@ func (d *Discovery) newEndpointSlicesByNodeInformer(plw *cache.ListWatch, object
return nodes, nil
}
return cache.NewSharedIndexInformer(plw, object, resyncPeriod, indexers)
return cache.NewSharedIndexInformer(plw, object, resyncDisabled, indexers)
}
func checkDiscoveryV1Supported(client kubernetes.Interface) (bool, error) {

9
discovery/kubernetes/node.go
View file

@ -55,7 +55,7 @@ func NewNode(l log.Logger, inf cache.SharedInformer) *Node {
l = log.NewNopLogger()
}
n := &Node{logger: l, informer: inf, store: inf.GetStore(), queue: workqueue.NewNamed("node")}
n.informer.AddEventHandler(cache.ResourceEventHandlerFuncs{
_, err := n.informer.AddEventHandler(cache.ResourceEventHandlerFuncs{
AddFunc: func(o interface{}) {
nodeAddCount.Inc()
n.enqueue(o)
@ -69,6 +69,9 @@ func NewNode(l log.Logger, inf cache.SharedInformer) *Node {
n.enqueue(o)
},
})
if err != nil {
level.Error(l).Log("msg", "Error adding nodes event handler.", "err", err)
}
return n
}
@ -93,7 +96,7 @@ func (n *Node) Run(ctx context.Context, ch chan<- []*targetgroup.Group) {
}
go func() {
for n.process(ctx, ch) {
for n.process(ctx, ch) { // nolint:revive
}
}()
@ -206,7 +209,7 @@ func (n *Node) buildNode(node *apiv1.Node) *targetgroup.Group {
return tg
}
// nodeAddresses returns the provided node's address, based on the priority:
// nodeAddress returns the provided node's address, based on the priority:
// 1. NodeInternalIP
// 2. NodeInternalDNS
// 3. NodeExternalIP

39
discovery/kubernetes/pod.go
View file

@ -65,7 +65,7 @@ func NewPod(l log.Logger, pods cache.SharedIndexInformer, nodes cache.SharedInfo
logger: l,
queue: workqueue.NewNamed("pod"),
}
p.podInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
_, err := p.podInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
AddFunc: func(o interface{}) {
podAddCount.Inc()
p.enqueue(o)
@ -79,9 +79,12 @@ func NewPod(l log.Logger, pods cache.SharedIndexInformer, nodes cache.SharedInfo
p.enqueue(o)
},
})
if err != nil {
level.Error(l).Log("msg", "Error adding pods event handler.", "err", err)
}
if p.withNodeMetadata {
p.nodeInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
_, err = p.nodeInf.AddEventHandler(cache.ResourceEventHandlerFuncs{
AddFunc: func(o interface{}) {
node := o.(*apiv1.Node)
p.enqueuePodsForNode(node.Name)
@ -95,6 +98,9 @@ func NewPod(l log.Logger, pods cache.SharedIndexInformer, nodes cache.SharedInfo
p.enqueuePodsForNode(node.Name)
},
})
if err != nil {
level.Error(l).Log("msg", "Error adding pods event handler.", "err", err)
}
}
return p
@ -126,7 +132,7 @@ func (p *Pod) Run(ctx context.Context, ch chan<- []*targetgroup.Group) {
}
go func() {
for p.process(ctx, ch) {
for p.process(ctx, ch) { // nolint:revive
}
}()
@ -177,6 +183,7 @@ const (
podNameLabel = metaLabelPrefix + "pod_name"
podIPLabel = metaLabelPrefix + "pod_ip"
podContainerNameLabel = metaLabelPrefix + "pod_container_name"
podContainerIDLabel = metaLabelPrefix + "pod_container_id"
podContainerImageLabel = metaLabelPrefix + "pod_container_image"
podContainerPortNameLabel = metaLabelPrefix + "pod_container_port_name"
podContainerPortNumberLabel = metaLabelPrefix + "pod_container_port_number"
@ -242,6 +249,24 @@ func podLabels(pod *apiv1.Pod) model.LabelSet {
return ls
}
func (p *Pod) findPodContainerStatus(statuses *[]apiv1.ContainerStatus, containerName string) (*apiv1.ContainerStatus, error) {
for _, s := range *statuses {
if s.Name == containerName {
return &s, nil
}
}
return nil, fmt.Errorf("cannot find container with name %v", containerName)
}
func (p *Pod) findPodContainerID(statuses *[]apiv1.ContainerStatus, containerName string) string {
cStatus, err := p.findPodContainerStatus(statuses, containerName)
if err != nil {
level.Debug(p.logger).Log("msg", "cannot find container ID", "err", err)
return ""
}
return cStatus.ContainerID
}
func (p *Pod) buildPod(pod *apiv1.Pod) *targetgroup.Group {
tg := &targetgroup.Group{
Source: podSource(pod),
@ -261,6 +286,12 @@ func (p *Pod) buildPod(pod *apiv1.Pod) *targetgroup.Group {
for i, c := range containers {
isInit := i >= len(pod.Spec.Containers)
cStatuses := &pod.Status.ContainerStatuses
if isInit {
cStatuses = &pod.Status.InitContainerStatuses
}
cID := p.findPodContainerID(cStatuses, c.Name)
// If no ports are defined for the container, create an anonymous
// target per container.
if len(c.Ports) == 0 {
@ -269,6 +300,7 @@ func (p *Pod) buildPod(pod *apiv1.Pod) *targetgroup.Group {
tg.Targets = append(tg.Targets, model.LabelSet{
model.AddressLabel: lv(pod.Status.PodIP),
podContainerNameLabel: lv(c.Name),
podContainerIDLabel: lv(cID),
podContainerImageLabel: lv(c.Image),
podContainerIsInit: lv(strconv.FormatBool(isInit)),
})
@ -282,6 +314,7 @@ func (p *Pod) buildPod(pod *apiv1.Pod) *targetgroup.Group {
tg.Targets = append(tg.Targets, model.LabelSet{
model.AddressLabel: lv(addr),
podContainerNameLabel: lv(c.Name),
podContainerIDLabel: lv(cID),
podContainerImageLabel: lv(c.Image),
podContainerPortNumberLabel: lv(ports),
podContainerPortNameLabel: lv(port.Name),

33
discovery/kubernetes/pod_test.go
View file

@ -81,6 +81,16 @@ func makeMultiPortPods() *v1.Pod {
Status: v1.ConditionTrue,
},
},
ContainerStatuses: []v1.ContainerStatus{
{
Name: "testcontainer0",
ContainerID: "docker://a1b2c3d4e5f6",
},
{
Name: "testcontainer1",
ContainerID: "containerd://6f5e4d3c2b1a",
},
},
},
}
}
@ -118,6 +128,12 @@ func makePods() *v1.Pod {
Status: v1.ConditionTrue,
},
},
ContainerStatuses: []v1.ContainerStatus{
{
Name: "testcontainer",
ContainerID: "docker://a1b2c3d4e5f6",
},
},
},
}
}
@ -162,6 +178,18 @@ func makeInitContainerPods() *v1.Pod {
Status: v1.ConditionFalse,
},
},
ContainerStatuses: []v1.ContainerStatus{
{
Name: "testcontainer",
ContainerID: "docker://a1b2c3d4e5f6",
},
},
InitContainerStatuses: []v1.ContainerStatus{
{
Name: "initcontainer",
ContainerID: "containerd://6f5e4d3c2b1a",
},
},
},
}
}
@ -179,6 +207,7 @@ func expectedPodTargetGroups(ns string) map[string]*targetgroup.Group {
"__meta_kubernetes_pod_container_port_number": "9000",
"__meta_kubernetes_pod_container_port_protocol": "TCP",
"__meta_kubernetes_pod_container_init": "false",
"__meta_kubernetes_pod_container_id": "docker://a1b2c3d4e5f6",
},
},
Labels: model.LabelSet{
@ -230,6 +259,7 @@ func TestPodDiscoveryBeforeRun(t *testing.T) {
"__meta_kubernetes_pod_container_port_number": "9000",
"__meta_kubernetes_pod_container_port_protocol": "TCP",
"__meta_kubernetes_pod_container_init": "false",
"__meta_kubernetes_pod_container_id": "docker://a1b2c3d4e5f6",
},
{
"__address__": "1.2.3.4:9001",
@ -239,12 +269,14 @@ func TestPodDiscoveryBeforeRun(t *testing.T) {
"__meta_kubernetes_pod_container_port_number": "9001",
"__meta_kubernetes_pod_container_port_protocol": "UDP",
"__meta_kubernetes_pod_container_init": "false",
"__meta_kubernetes_pod_container_id": "docker://a1b2c3d4e5f6",
},
{
"__address__": "1.2.3.4",
"__meta_kubernetes_pod_container_name": "testcontainer1",
"__meta_kubernetes_pod_container_image": "testcontainer1:latest",
"__meta_kubernetes_pod_container_init": "false",
"__meta_kubernetes_pod_container_id": "containerd://6f5e4d3c2b1a",
},
},
Labels: model.LabelSet{
@ -280,6 +312,7 @@ func TestPodDiscoveryInitContainer(t *testing.T) {
"__meta_kubernetes_pod_container_name": "initcontainer",
"__meta_kubernetes_pod_container_image": "initcontainer:latest",
"__meta_kubernetes_pod_container_init": "true",
"__meta_kubernetes_pod_container_id": "containerd://6f5e4d3c2b1a",
})
expected[key].Labels["__meta_kubernetes_pod_phase"] = "Pending"
expected[key].Labels["__meta_kubernetes_pod_ready"] = "false"

7
discovery/kubernetes/service.go
View file

@ -51,7 +51,7 @@ func NewService(l log.Logger, inf cache.SharedInformer) *Service {
l = log.NewNopLogger()
}
s := &Service{logger: l, informer: inf, store: inf.GetStore(), queue: workqueue.NewNamed("service")}
s.informer.AddEventHandler(cache.ResourceEventHandlerFuncs{
_, err := s.informer.AddEventHandler(cache.ResourceEventHandlerFuncs{
AddFunc: func(o interface{}) {
svcAddCount.Inc()
s.enqueue(o)
@ -65,6 +65,9 @@ func NewService(l log.Logger, inf cache.SharedInformer) *Service {
s.enqueue(o)
},
})
if err != nil {
level.Error(l).Log("msg", "Error adding services event handler.", "err", err)
}
return s
}
@ -89,7 +92,7 @@ func (s *Service) Run(ctx context.Context, ch chan<- []*targetgroup.Group) {
}
go func() {
for s.process(ctx, ch) {
for s.process(ctx, ch) { // nolint:revive
}
}()

Some files were not shown because too many files have changed in this diff Show more