Compare commits

..

1 Commits

Author SHA1 Message Date
Haley Wang
23c615f046 docs: add HA options 2024-06-17 22:00:45 +08:00
1522 changed files with 41536 additions and 68782 deletions

View File

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

View File

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

View File

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

103
.github/workflows/codeql-analysis.yml vendored Normal file
View File

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

View File

@@ -1,25 +1,29 @@
name: main
on:
push:
branches:
- cluster
- master
paths:
- '**.go'
- cluster
paths-ignore:
- "docs/**"
- "**.md"
- "dashboards/**"
- "deployment/**.yml"
pull_request:
branches:
- cluster
- master
paths:
- '**.go'
- cluster
paths-ignore:
- "docs/**"
- "**.md"
- "dashboards/**"
- "deployment/**.yml"
permissions:
contents: read
concurrency:
cancel-in-progress: true
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
cancel-in-progress: true
jobs:
lint:
@@ -33,16 +37,16 @@ jobs:
id: go
uses: actions/setup-go@v5
with:
cache: false
go-version: stable
cache: false
- name: Cache Go artifacts
uses: actions/cache@v4
with:
path: |
~/.cache/go-build
~/go/bin
~/go/pkg/mod
~/go/bin
key: go-artifacts-${{ runner.os }}-check-all-${{ steps.go.outputs.go-version }}-${{ hashFiles('go.sum', 'Makefile', 'app/**/Makefile') }}
restore-keys: go-artifacts-${{ runner.os }}-check-all-
@@ -51,18 +55,10 @@ jobs:
make check-all
git diff --exit-code
test:
name: test
build:
needs: lint
name: build
runs-on: ubuntu-latest
strategy:
matrix:
scenario:
- 'test-full'
- 'test-full-386'
- 'test-pure'
steps:
- name: Code checkout
uses: actions/checkout@v4
@@ -71,20 +67,51 @@ jobs:
id: go
uses: actions/setup-go@v5
with:
cache: false
go-version: stable
cache: false
- name: Cache Go artifacts
uses: actions/cache@v4
with:
path: |
~/.cache/go-build
~/go/bin
~/go/pkg/mod
~/go/bin
key: go-artifacts-${{ runner.os }}-crossbuild-${{ steps.go.outputs.go-version }}-${{ hashFiles('go.sum', 'Makefile', 'app/**/Makefile') }}
restore-keys: go-artifacts-${{ runner.os }}-crossbuild-
- name: Build
run: make crossbuild
test:
needs: lint
strategy:
matrix:
scenario: ["test-full", "test-pure", "test-full-386"]
name: test
runs-on: ubuntu-latest
steps:
- name: Code checkout
uses: actions/checkout@v4
- name: Setup Go
id: go
uses: actions/setup-go@v5
with:
go-version: stable
cache: false
- name: Cache Go artifacts
uses: actions/cache@v4
with:
path: |
~/.cache/go-build
~/go/pkg/mod
~/go/bin
key: go-artifacts-${{ runner.os }}-${{ matrix.scenario }}-${{ steps.go.outputs.go-version }}-${{ hashFiles('go.sum', 'Makefile', 'app/**/Makefile') }}
restore-keys: go-artifacts-${{ runner.os }}-${{ matrix.scenario }}-
- name: Run tests
- name: run tests
run: make ${{ matrix.scenario}}
- name: Publish coverage

View File

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

5
.gitignore vendored
View File

@@ -7,7 +7,6 @@
.vscode
*.test
*.swp
/vmdocs
/gocache-for-docker
/victoria-logs-data
/victoria-metrics-data
@@ -23,6 +22,4 @@ Gemfile.lock
/_site
_site
*.tmp
/docs/.jekyll-metadata
coverage.txt
cspell.json
/docs/.jekyll-metadata

120
CODE_OF_CONDUCT_RU.md Normal file
View File

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

View File

@@ -17,7 +17,6 @@ GO_BUILDINFO = -X '$(PKG_PREFIX)/lib/buildinfo.Version=$(APP_NAME)-$(DATEINFO_TA
.PHONY: $(MAKECMDGOALS)
include app/*/Makefile
include cspell/Makefile
include docs/Makefile
include deployment/*/Makefile
include dashboards/Makefile
@@ -180,8 +179,6 @@ vet:
check-all: fmt vet golangci-lint govulncheck
clean-checkers: remove-golangci-lint remove-govulncheck
test:
go test ./lib/... ./app/...
@@ -234,10 +231,7 @@ golangci-lint: install-golangci-lint
golangci-lint run
install-golangci-lint:
which golangci-lint || curl -sSfL https://raw.githubusercontent.com/golangci/golangci-lint/master/install.sh | sh -s -- -b $(shell go env GOPATH)/bin v1.60.1
remove-golangci-lint:
rm -rf `which golangci-lint`
which golangci-lint || curl -sSfL https://raw.githubusercontent.com/golangci/golangci-lint/master/install.sh | sh -s -- -b $(shell go env GOPATH)/bin v1.59.1
govulncheck: install-govulncheck
govulncheck ./...
@@ -245,11 +239,37 @@ govulncheck: install-govulncheck
install-govulncheck:
which govulncheck || go install golang.org/x/vuln/cmd/govulncheck@latest
remove-govulncheck:
rm -rf `which govulncheck`
install-wwhrd:
which wwhrd || go install github.com/frapposelli/wwhrd@latest
check-licenses: install-wwhrd
wwhrd check -f .wwhrd.yml
copy-docs:
# The 'printf' function is used instead of 'echo' or 'echo -e' to handle line breaks (e.g. '\n') in the same way on different operating systems (MacOS/Ubuntu Linux/Arch Linux) and their shells (bash/sh/zsh/fish).
# For details, see https://github.com/VictoriaMetrics/VictoriaMetrics/pull/4548#issue-1782796419 and https://stackoverflow.com/questions/8467424/echo-newline-in-bash-prints-literal-n
echo "---" > ${DST}
@if [ ${ORDER} -ne 0 ]; then \
echo "sort: ${ORDER}" >> ${DST}; \
echo "weight: ${ORDER}" >> ${DST}; \
printf "menu:\n docs:\n parent: 'victoriametrics'\n weight: ${ORDER}\n" >> ${DST}; \
fi
echo "title: ${TITLE}" >> ${DST}
@if [ ${OLD_URL} ]; then \
printf "aliases:\n - ${OLD_URL}\n" >> ${DST}; \
fi
echo "---" >> ${DST}
cat ${SRC} >> ${DST}
sed -i='.tmp' 's/<img src=\"docs\//<img src=\"/' ${DST}
rm -rf docs/*.tmp
# Copies docs for all components and adds the order/weight tag, title, menu position and alias with the backward compatible link for the old site.
# For ORDER=0 it adds no order tag/weight tag.
# FOR OLD_URL - relative link, used for backward compatibility with the link from documentation based on GitHub pages (old one)
# FOR OLD_URL='' it adds no alias, it should be empty for every new page, don't change it for already existing links.
# Images starting with <img src="docs/ are replaced with <img src="
# Cluster docs are supposed to be ordered as 2nd.
# The rest of docs is ordered manually.
docs-sync:
SRC=README.md DST=docs/Cluster-VictoriaMetrics.md OLD_URL='/Cluster-VictoriaMetrics.html' ORDER=2 TITLE='Cluster version' $(MAKE) copy-docs

1938
README.md

File diff suppressed because it is too large Load Diff

View File

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

View File

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

View File

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

View File

@@ -57,12 +57,6 @@ func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
fmt.Fprintf(w, `{}`)
return true
}
if strings.HasPrefix(path, "/logstash") || strings.HasPrefix(path, "/_logstash") {
// Return fake response for Logstash APIs requests.
// See: https://www.elastic.co/guide/en/elasticsearch/reference/8.8/logstash-apis.html
fmt.Fprintf(w, `{}`)
return true
}
switch path {
case "/":
switch r.Method {
@@ -103,10 +97,12 @@ func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
httpserver.Errorf(w, r, "%s", err)
return true
}
lmp := cp.NewLogMessageProcessor()
lr := logstorage.GetLogRows(cp.StreamFields, cp.IgnoreFields)
processLogMessage := cp.GetProcessLogMessageFunc(lr)
isGzip := r.Header.Get("Content-Encoding") == "gzip"
n, err := readBulkRequest(r.Body, isGzip, cp.TimeField, cp.MsgField, lmp)
lmp.MustClose()
n, err := readBulkRequest(r.Body, isGzip, cp.TimeField, cp.MsgField, processLogMessage)
vlstorage.MustAddRows(lr)
logstorage.PutLogRows(lr)
if err != nil {
logger.Warnf("cannot decode log message #%d in /_bulk request: %s, stream fields: %s", n, err, cp.StreamFields)
return true
@@ -135,7 +131,9 @@ var (
bulkRequestDuration = metrics.NewHistogram(`vl_http_request_duration_seconds{path="/insert/elasticsearch/_bulk"}`)
)
func readBulkRequest(r io.Reader, isGzip bool, timeField, msgField string, lmp insertutils.LogMessageProcessor) (int, error) {
func readBulkRequest(r io.Reader, isGzip bool, timeField, msgField string,
processLogMessage func(timestamp int64, fields []logstorage.Field),
) (int, error) {
// See https://www.elastic.co/guide/en/elasticsearch/reference/current/docs-bulk.html
if isGzip {
@@ -160,7 +158,7 @@ func readBulkRequest(r io.Reader, isGzip bool, timeField, msgField string, lmp i
n := 0
nCheckpoint := 0
for {
ok, err := readBulkLine(sc, timeField, msgField, lmp)
ok, err := readBulkLine(sc, timeField, msgField, processLogMessage)
wcr.DecConcurrency()
if err != nil || !ok {
rowsIngestedTotal.Add(n - nCheckpoint)
@@ -176,7 +174,9 @@ func readBulkRequest(r io.Reader, isGzip bool, timeField, msgField string, lmp i
var lineBufferPool bytesutil.ByteBufferPool
func readBulkLine(sc *bufio.Scanner, timeField, msgField string, lmp insertutils.LogMessageProcessor) (bool, error) {
func readBulkLine(sc *bufio.Scanner, timeField, msgField string,
processLogMessage func(timestamp int64, fields []logstorage.Field),
) (bool, error) {
var line []byte
// Read the command, must be "create" or "index"
@@ -222,7 +222,7 @@ func readBulkLine(sc *bufio.Scanner, timeField, msgField string, lmp insertutils
ts = time.Now().UnixNano()
}
logstorage.RenameField(p.Fields, msgField, "_msg")
lmp.AddRow(ts, p.Fields)
processLogMessage(ts, p.Fields)
logstorage.PutJSONParser(p)
return true, nil

View File

@@ -4,18 +4,22 @@ import (
"bytes"
"compress/gzip"
"fmt"
"reflect"
"testing"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
)
func TestReadBulkRequest_Failure(t *testing.T) {
func TestReadBulkRequestFailure(t *testing.T) {
f := func(data string) {
t.Helper()
tlp := &insertutils.TestLogMessageProcessor{}
processLogMessage := func(timestamp int64, fields []logstorage.Field) {
t.Fatalf("unexpected call to processLogMessage with timestamp=%d, fields=%s", timestamp, fields)
}
r := bytes.NewBufferString(data)
rows, err := readBulkRequest(r, false, "_time", "_msg", tlp)
rows, err := readBulkRequest(r, false, "_time", "_msg", processLogMessage)
if err == nil {
t.Fatalf("expecting non-empty error")
}
@@ -32,38 +36,52 @@ func TestReadBulkRequest_Failure(t *testing.T) {
foobar`)
}
func TestReadBulkRequest_Success(t *testing.T) {
func TestReadBulkRequestSuccess(t *testing.T) {
f := func(data, timeField, msgField string, rowsExpected int, timestampsExpected []int64, resultExpected string) {
t.Helper()
tlp := &insertutils.TestLogMessageProcessor{}
var timestamps []int64
var result string
processLogMessage := func(timestamp int64, fields []logstorage.Field) {
timestamps = append(timestamps, timestamp)
result += string(logstorage.MarshalFieldsToJSON(nil, fields)) + "\n"
}
// Read the request without compression
r := bytes.NewBufferString(data)
rows, err := readBulkRequest(r, false, timeField, msgField, tlp)
rows, err := readBulkRequest(r, false, timeField, msgField, processLogMessage)
if err != nil {
t.Fatalf("unexpected error: %s", err)
}
if rows != rowsExpected {
t.Fatalf("unexpected rows read; got %d; want %d", rows, rowsExpected)
}
if err := tlp.Verify(rowsExpected, timestampsExpected, resultExpected); err != nil {
t.Fatal(err)
if !reflect.DeepEqual(timestamps, timestampsExpected) {
t.Fatalf("unexpected timestamps;\ngot\n%d\nwant\n%d", timestamps, timestampsExpected)
}
if result != resultExpected {
t.Fatalf("unexpected result;\ngot\n%s\nwant\n%s", result, resultExpected)
}
// Read the request with compression
tlp = &insertutils.TestLogMessageProcessor{}
timestamps = nil
result = ""
compressedData := compressData(data)
r = bytes.NewBufferString(compressedData)
rows, err = readBulkRequest(r, true, timeField, msgField, tlp)
rows, err = readBulkRequest(r, true, timeField, msgField, processLogMessage)
if err != nil {
t.Fatalf("unexpected error: %s", err)
}
if rows != rowsExpected {
t.Fatalf("unexpected rows read; got %d; want %d", rows, rowsExpected)
}
if err := tlp.Verify(rowsExpected, timestampsExpected, resultExpected); err != nil {
t.Fatalf("verification failure after compression: %s", err)
if !reflect.DeepEqual(timestamps, timestampsExpected) {
t.Fatalf("unexpected timestamps;\ngot\n%d\nwant\n%d", timestamps, timestampsExpected)
}
if result != resultExpected {
t.Fatalf("unexpected result;\ngot\n%s\nwant\n%s", result, resultExpected)
}
}
@@ -86,7 +104,8 @@ func TestReadBulkRequest_Success(t *testing.T) {
timestampsExpected := []int64{1686026891735000000, 1686026892735000000, 1686026893735000000}
resultExpected := `{"@timestamp":"","log.offset":"71770","log.file.path":"/var/log/auth.log","_msg":"foobar"}
{"@timestamp":"","_msg":"baz"}
{"_msg":"xyz","@timestamp":"","x":"y"}`
{"_msg":"xyz","@timestamp":"","x":"y"}
`
f(data, timeField, msgField, rowsExpected, timestampsExpected, resultExpected)
}

View File

@@ -5,8 +5,8 @@ import (
"fmt"
"testing"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
)
func BenchmarkReadBulkRequest(b *testing.B) {
@@ -33,7 +33,7 @@ func benchmarkReadBulkRequest(b *testing.B, isGzip bool) {
timeField := "@timestamp"
msgField := "message"
blp := &insertutils.BenchmarkLogMessageProcessor{}
processLogMessage := func(_ int64, _ []logstorage.Field) {}
b.ReportAllocs()
b.SetBytes(int64(len(data)))
@@ -41,7 +41,7 @@ func benchmarkReadBulkRequest(b *testing.B, isGzip bool) {
r := &bytes.Reader{}
for pb.Next() {
r.Reset(dataBytes)
_, err := readBulkRequest(r, isGzip, timeField, msgField, blp)
_, err := readBulkRequest(r, isGzip, timeField, msgField, processLogMessage)
if err != nil {
panic(fmt.Errorf("unexpected error: %w", err))
}

View File

@@ -2,8 +2,6 @@ package insertutils
import (
"net/http"
"sync"
"time"
"github.com/VictoriaMetrics/metrics"
@@ -12,7 +10,6 @@ import (
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/timeutil"
)
// CommonParams contains common HTTP parameters used by log ingestion APIs.
@@ -91,109 +88,29 @@ func GetCommonParamsForSyslog(tenantID logstorage.TenantID) *CommonParams {
return cp
}
// LogMessageProcessor is an interface for log message processors.
type LogMessageProcessor interface {
// AddRow must add row to the LogMessageProcessor with the given timestamp and the given fields.
//
// The LogMessageProcessor implementation cannot hold references to fields, since the caller can re-use them.
AddRow(timestamp int64, fields []logstorage.Field)
// MustClose() must flush all the remaining fields and free up resources occupied by LogMessageProcessor.
MustClose()
}
type logMessageProcessor struct {
mu sync.Mutex
wg sync.WaitGroup
stopCh chan struct{}
lastFlushTime time.Time
cp *CommonParams
lr *logstorage.LogRows
}
func (lmp *logMessageProcessor) initPeriodicFlush() {
lmp.lastFlushTime = time.Now()
lmp.wg.Add(1)
go func() {
defer lmp.wg.Done()
d := timeutil.AddJitterToDuration(time.Second)
ticker := time.NewTicker(d)
defer ticker.Stop()
for {
select {
case <-lmp.stopCh:
return
case <-ticker.C:
lmp.mu.Lock()
if time.Since(lmp.lastFlushTime) >= d {
lmp.flushLocked()
}
lmp.mu.Unlock()
}
// GetProcessLogMessageFunc returns a function, which adds parsed log messages to lr.
func (cp *CommonParams) GetProcessLogMessageFunc(lr *logstorage.LogRows) func(timestamp int64, fields []logstorage.Field) {
return func(timestamp int64, fields []logstorage.Field) {
if len(fields) > *MaxFieldsPerLine {
rf := logstorage.RowFormatter(fields)
logger.Warnf("dropping log line with %d fields; it exceeds -insert.maxFieldsPerLine=%d; %s", len(fields), *MaxFieldsPerLine, rf)
rowsDroppedTotalTooManyFields.Inc()
return
}
}()
}
// AddRow adds new log message to lmp with the given timestamp and fields.
func (lmp *logMessageProcessor) AddRow(timestamp int64, fields []logstorage.Field) {
lmp.mu.Lock()
defer lmp.mu.Unlock()
if len(fields) > *MaxFieldsPerLine {
rf := logstorage.RowFormatter(fields)
logger.Warnf("dropping log line with %d fields; it exceeds -insert.maxFieldsPerLine=%d; %s", len(fields), *MaxFieldsPerLine, rf)
rowsDroppedTotalTooManyFields.Inc()
return
lr.MustAdd(cp.TenantID, timestamp, fields)
if cp.Debug {
s := lr.GetRowString(0)
lr.ResetKeepSettings()
logger.Infof("remoteAddr=%s; requestURI=%s; ignoring log entry because of `debug` query arg: %s", cp.DebugRemoteAddr, cp.DebugRequestURI, s)
rowsDroppedTotalDebug.Inc()
return
}
if lr.NeedFlush() {
vlstorage.MustAddRows(lr)
lr.ResetKeepSettings()
}
}
lmp.lr.MustAdd(lmp.cp.TenantID, timestamp, fields)
if lmp.cp.Debug {
s := lmp.lr.GetRowString(0)
lmp.lr.ResetKeepSettings()
logger.Infof("remoteAddr=%s; requestURI=%s; ignoring log entry because of `debug` query arg: %s", lmp.cp.DebugRemoteAddr, lmp.cp.DebugRequestURI, s)
rowsDroppedTotalDebug.Inc()
return
}
if lmp.lr.NeedFlush() {
lmp.flushLocked()
}
}
// flushLocked must be called under locked lmp.mu.
func (lmp *logMessageProcessor) flushLocked() {
lmp.lastFlushTime = time.Now()
vlstorage.MustAddRows(lmp.lr)
lmp.lr.ResetKeepSettings()
}
// MustClose flushes the remaining data to the underlying storage and closes lmp.
func (lmp *logMessageProcessor) MustClose() {
close(lmp.stopCh)
lmp.wg.Wait()
lmp.flushLocked()
logstorage.PutLogRows(lmp.lr)
lmp.lr = nil
}
// NewLogMessageProcessor returns new LogMessageProcessor for the given cp.
//
// MustClose() must be called on the returned LogMessageProcessor when it is no longer needed.
func (cp *CommonParams) NewLogMessageProcessor() LogMessageProcessor {
lr := logstorage.GetLogRows(cp.StreamFields, cp.IgnoreFields)
lmp := &logMessageProcessor{
cp: cp,
lr: lr,
stopCh: make(chan struct{}),
}
lmp.initPeriodicFlush()
return lmp
}
var rowsDroppedTotalDebug = metrics.NewCounter(`vl_rows_dropped_total{reason="debug"}`)

View File

@@ -1,53 +0,0 @@
package insertutils
import (
"fmt"
"reflect"
"strings"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
)
// TestLogMessageProcessor implements LogMessageProcessor for testing.
type TestLogMessageProcessor struct {
timestamps []int64
rows []string
}
// AddRow adds row with the given timestamp and fields to tlp
func (tlp *TestLogMessageProcessor) AddRow(timestamp int64, fields []logstorage.Field) {
tlp.timestamps = append(tlp.timestamps, timestamp)
tlp.rows = append(tlp.rows, string(logstorage.MarshalFieldsToJSON(nil, fields)))
}
// MustClose closes tlp.
func (tlp *TestLogMessageProcessor) MustClose() {
}
// Verify verifies the number of rows, timestamps and results after AddRow calls.
func (tlp *TestLogMessageProcessor) Verify(rowsExpected int, timestampsExpected []int64, resultExpected string) error {
result := strings.Join(tlp.rows, "\n")
if len(tlp.rows) != rowsExpected {
return fmt.Errorf("unexpected rows read; got %d; want %d;\nrows read:\n%s\nrows wanted\n%s", len(tlp.rows), rowsExpected, result, resultExpected)
}
if !reflect.DeepEqual(tlp.timestamps, timestampsExpected) {
return fmt.Errorf("unexpected timestamps;\ngot\n%d\nwant\n%d", tlp.timestamps, timestampsExpected)
}
if result != resultExpected {
return fmt.Errorf("unexpected result;\ngot\n%s\nwant\n%s", result, resultExpected)
}
return nil
}
// BenchmarkLogMessageProcessor implements LogMessageProcessor for benchmarks.
type BenchmarkLogMessageProcessor struct{}
// AddRow implements LogMessageProcessor interface.
func (blp *BenchmarkLogMessageProcessor) AddRow(_ int64, _ []logstorage.Field) {
}
// MustClose implements LogMessageProcessor interface.
func (blp *BenchmarkLogMessageProcessor) MustClose() {
}

View File

@@ -7,24 +7,24 @@ import (
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
)
// ExtractTimestampRFC3339NanoFromFields extracts RFC3339 timestamp in nanoseconds from the field with the name timeField at fields.
// ExtractTimestampISO8601FromFields extracts ISO8601 timestamp in nanoseconds from the field with the name timeField at fields.
//
// The value for the timeField is set to empty string after returning from the function,
// so it could be ignored during data ingestion.
//
// The current timestamp is returned if fields do not contain a field with timeField name or if the timeField value is empty.
func ExtractTimestampRFC3339NanoFromFields(timeField string, fields []logstorage.Field) (int64, error) {
func ExtractTimestampISO8601FromFields(timeField string, fields []logstorage.Field) (int64, error) {
for i := range fields {
f := &fields[i]
if f.Name != timeField {
continue
}
if f.Value == "" || f.Value == "0" {
return time.Now().UnixNano(), nil
}
nsecs, ok := logstorage.TryParseTimestampRFC3339Nano(f.Value)
nsecs, ok := logstorage.TryParseTimestampISO8601(f.Value)
if !ok {
return 0, fmt.Errorf("cannot unmarshal rfc3339 timestamp from %s=%q", timeField, f.Value)
if f.Value == "0" || f.Value == "" {
return time.Now().UnixNano(), nil
}
return time.Now().UnixNano(), fmt.Errorf("cannot unmarshal iso8601 timestamp from %s=%q", timeField, f.Value)
}
f.Value = ""
return nsecs, nil

View File

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

View File

@@ -4,7 +4,6 @@ import (
"bufio"
"errors"
"fmt"
"io"
"net/http"
"time"
@@ -40,6 +39,8 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) {
httpserver.Errorf(w, r, "%s", err)
return
}
lr := logstorage.GetLogRows(cp.StreamFields, cp.IgnoreFields)
processLogMessage := cp.GetProcessLogMessageFunc(lr)
reader := r.Body
if r.Header.Get("Content-Encoding") == "gzip" {
@@ -52,22 +53,7 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) {
reader = zr
}
lmp := cp.NewLogMessageProcessor()
err = processStreamInternal(reader, cp.TimeField, cp.MsgField, lmp)
lmp.MustClose()
if err != nil {
logger.Errorf("jsonline: %s", err)
} else {
// update requestDuration only for successfully parsed requests.
// There is no need in updating requestDuration for request errors,
// since their timings are usually much smaller than the timing for successful request parsing.
requestDuration.UpdateDuration(startTime)
}
}
func processStreamInternal(r io.Reader, timeField, msgField string, lmp insertutils.LogMessageProcessor) error {
wcr := writeconcurrencylimiter.GetReader(r)
wcr := writeconcurrencylimiter.GetReader(reader)
defer writeconcurrencylimiter.PutReader(wcr)
lb := lineBufferPool.Get()
@@ -79,21 +65,30 @@ func processStreamInternal(r io.Reader, timeField, msgField string, lmp insertut
n := 0
for {
ok, err := readLine(sc, timeField, msgField, lmp)
ok, err := readLine(sc, cp.TimeField, cp.MsgField, processLogMessage)
wcr.DecConcurrency()
if err != nil {
errorsTotal.Inc()
return fmt.Errorf("cannot read line #%d in /jsonline request: %s", n, err)
logger.Errorf("cannot read line #%d in /jsonline request: %s", n, err)
break
}
if !ok {
return nil
break
}
n++
rowsIngestedTotal.Inc()
}
vlstorage.MustAddRows(lr)
logstorage.PutLogRows(lr)
// update requestDuration only for successfully parsed requests.
// There is no need in updating requestDuration for request errors,
// since their timings are usually much smaller than the timing for successful request parsing.
requestDuration.UpdateDuration(startTime)
}
func readLine(sc *bufio.Scanner, timeField, msgField string, lmp insertutils.LogMessageProcessor) (bool, error) {
func readLine(sc *bufio.Scanner, timeField, msgField string, processLogMessage func(timestamp int64, fields []logstorage.Field)) (bool, error) {
var line []byte
for len(line) == 0 {
if !sc.Scan() {
@@ -112,12 +107,12 @@ func readLine(sc *bufio.Scanner, timeField, msgField string, lmp insertutils.Log
if err := p.ParseLogMessage(line); err != nil {
return false, fmt.Errorf("cannot parse json-encoded log entry: %w", err)
}
ts, err := insertutils.ExtractTimestampRFC3339NanoFromFields(timeField, p.Fields)
ts, err := insertutils.ExtractTimestampISO8601FromFields(timeField, p.Fields)
if err != nil {
return false, fmt.Errorf("cannot get timestamp: %w", err)
}
logstorage.RenameField(p.Fields, msgField, "_msg")
lmp.AddRow(ts, p.Fields)
processLogMessage(ts, p.Fields)
logstorage.PutJSONParser(p)
return true, nil

View File

@@ -1,24 +1,47 @@
package jsonline
import (
"bufio"
"bytes"
"reflect"
"testing"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
)
func TestProcessStreamInternal_Success(t *testing.T) {
func TestReadLine_Success(t *testing.T) {
f := func(data, timeField, msgField string, rowsExpected int, timestampsExpected []int64, resultExpected string) {
t.Helper()
tlp := &insertutils.TestLogMessageProcessor{}
r := bytes.NewBufferString(data)
if err := processStreamInternal(r, timeField, msgField, tlp); err != nil {
t.Fatalf("unexpected error: %s", err)
var timestamps []int64
var result string
processLogMessage := func(timestamp int64, fields []logstorage.Field) {
timestamps = append(timestamps, timestamp)
result += string(logstorage.MarshalFieldsToJSON(nil, fields)) + "\n"
}
if err := tlp.Verify(rowsExpected, timestampsExpected, resultExpected); err != nil {
t.Fatal(err)
r := bytes.NewBufferString(data)
sc := bufio.NewScanner(r)
rows := 0
for {
ok, err := readLine(sc, timeField, msgField, processLogMessage)
if err != nil {
t.Fatalf("unexpected error: %s", err)
}
if !ok {
break
}
rows++
}
if rows != rowsExpected {
t.Fatalf("unexpected rows read; got %d; want %d", rows, rowsExpected)
}
if !reflect.DeepEqual(timestamps, timestampsExpected) {
t.Fatalf("unexpected timestamps;\ngot\n%d\nwant\n%d", timestamps, timestampsExpected)
}
if result != resultExpected {
t.Fatalf("unexpected result;\ngot\n%s\nwant\n%s", result, resultExpected)
}
}
@@ -32,24 +55,7 @@ func TestProcessStreamInternal_Success(t *testing.T) {
timestampsExpected := []int64{1686026891735000000, 1686026892735000000, 1686026893735000000}
resultExpected := `{"@timestamp":"","log.offset":"71770","log.file.path":"/var/log/auth.log","_msg":"foobar"}
{"@timestamp":"","_msg":"baz"}
{"_msg":"xyz","@timestamp":"","x":"y"}`
{"_msg":"xyz","@timestamp":"","x":"y"}
`
f(data, timeField, msgField, rowsExpected, timestampsExpected, resultExpected)
}
func TestProcessStreamInternal_Failure(t *testing.T) {
f := func(data string) {
t.Helper()
tlp := &insertutils.TestLogMessageProcessor{}
r := bytes.NewBufferString(data)
if err := processStreamInternal(r, "time", "", tlp); err == nil {
t.Fatalf("expecting non-nil error")
}
}
// invalid json
f("foobar")
// invalid timestamp field
f(`{"time":"foobar"}`)
}

View File

@@ -46,7 +46,7 @@ func getCommonParams(r *http.Request) (*insertutils.CommonParams, error) {
if cp.TenantID.AccountID == 0 && cp.TenantID.ProjectID == 0 {
org := r.Header.Get("X-Scope-OrgID")
if org != "" {
tenantID, err := logstorage.ParseTenantID(org)
tenantID, err := logstorage.GetTenantIDFromString(org)
if err != nil {
return nil, err
}

View File

@@ -8,7 +8,6 @@ import (
"strconv"
"time"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
@@ -52,9 +51,11 @@ func handleJSON(r *http.Request, w http.ResponseWriter) {
httpserver.Errorf(w, r, "%s", err)
return
}
lmp := cp.NewLogMessageProcessor()
n, err := parseJSONRequest(data, lmp)
lmp.MustClose()
lr := logstorage.GetLogRows(cp.StreamFields, cp.IgnoreFields)
processLogMessage := cp.GetProcessLogMessageFunc(lr)
n, err := parseJSONRequest(data, processLogMessage)
vlstorage.MustAddRows(lr)
logstorage.PutLogRows(lr)
if err != nil {
httpserver.Errorf(w, r, "cannot parse Loki json request: %s", err)
return
@@ -74,7 +75,7 @@ var (
requestJSONDuration = metrics.NewHistogram(`vl_http_request_duration_seconds{path="/insert/loki/api/v1/push",format="json"}`)
)
func parseJSONRequest(data []byte, lmp insertutils.LogMessageProcessor) (int, error) {
func parseJSONRequest(data []byte, processLogMessage func(timestamp int64, fields []logstorage.Field)) (int, error) {
p := parserPool.Get()
defer parserPool.Put(p)
v, err := p.ParseBytes(data)
@@ -167,7 +168,7 @@ func parseJSONRequest(data []byte, lmp insertutils.LogMessageProcessor) (int, er
Name: "_msg",
Value: bytesutil.ToUnsafeString(msg),
})
lmp.AddRow(ts, fields)
processLogMessage(ts, fields)
}
rowsIngested += len(lines)
}

View File

@@ -1,17 +1,19 @@
package loki
import (
"fmt"
"strings"
"testing"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
)
func TestParseJSONRequest_Failure(t *testing.T) {
func TestParseJSONRequestFailure(t *testing.T) {
f := func(s string) {
t.Helper()
tlp := &insertutils.TestLogMessageProcessor{}
n, err := parseJSONRequest([]byte(s), tlp)
n, err := parseJSONRequest([]byte(s), func(_ int64, _ []logstorage.Field) {
t.Fatalf("unexpected call to parseJSONRequest callback!")
})
if err == nil {
t.Fatalf("expecting non-nil error")
}
@@ -54,30 +56,39 @@ func TestParseJSONRequest_Failure(t *testing.T) {
f(`{"streams":[{"values":[["123",1234]]}]}`)
}
func TestParseJSONRequest_Success(t *testing.T) {
f := func(s string, timestampsExpected []int64, resultExpected string) {
func TestParseJSONRequestSuccess(t *testing.T) {
f := func(s string, resultExpected string) {
t.Helper()
tlp := &insertutils.TestLogMessageProcessor{}
n, err := parseJSONRequest([]byte(s), tlp)
var lines []string
n, err := parseJSONRequest([]byte(s), func(timestamp int64, fields []logstorage.Field) {
var a []string
for _, f := range fields {
a = append(a, f.String())
}
line := fmt.Sprintf("_time:%d %s", timestamp, strings.Join(a, " "))
lines = append(lines, line)
})
if err != nil {
t.Fatalf("unexpected error: %s", err)
}
if err := tlp.Verify(n, timestampsExpected, resultExpected); err != nil {
t.Fatal(err)
if n != len(lines) {
t.Fatalf("unexpected number of lines parsed; got %d; want %d", n, len(lines))
}
result := strings.Join(lines, "\n")
if result != resultExpected {
t.Fatalf("unexpected result;\ngot\n%s\nwant\n%s", result, resultExpected)
}
}
// Empty streams
f(`{"streams":[]}`, nil, ``)
f(`{"streams":[{"values":[]}]}`, nil, ``)
f(`{"streams":[{"stream":{},"values":[]}]}`, nil, ``)
f(`{"streams":[{"stream":{"foo":"bar"},"values":[]}]}`, nil, ``)
f(`{"streams":[]}`, ``)
f(`{"streams":[{"values":[]}]}`, ``)
f(`{"streams":[{"stream":{},"values":[]}]}`, ``)
f(`{"streams":[{"stream":{"foo":"bar"},"values":[]}]}`, ``)
// Empty stream labels
f(`{"streams":[{"values":[["1577836800000000001", "foo bar"]]}]}`, []int64{1577836800000000001}, `{"_msg":"foo bar"}`)
f(`{"streams":[{"stream":{},"values":[["1577836800000000001", "foo bar"]]}]}`, []int64{1577836800000000001}, `{"_msg":"foo bar"}`)
f(`{"streams":[{"values":[["1577836800000000001", "foo bar"]]}]}`, `_time:1577836800000000001 "_msg":"foo bar"`)
f(`{"streams":[{"stream":{},"values":[["1577836800000000001", "foo bar"]]}]}`, `_time:1577836800000000001 "_msg":"foo bar"`)
// Non-empty stream labels
f(`{"streams":[{"stream":{
@@ -87,9 +98,9 @@ func TestParseJSONRequest_Success(t *testing.T) {
["1577836800000000001", "foo bar"],
["1477836900005000002", "abc"],
["147.78369e9", "foobar"]
]}]}`, []int64{1577836800000000001, 1477836900005000002, 147783690000}, `{"label1":"value1","label2":"value2","_msg":"foo bar"}
{"label1":"value1","label2":"value2","_msg":"abc"}
{"label1":"value1","label2":"value2","_msg":"foobar"}`)
]}]}`, `_time:1577836800000000001 "label1":"value1" "label2":"value2" "_msg":"foo bar"
_time:1477836900005000002 "label1":"value1" "label2":"value2" "_msg":"abc"
_time:147783690000 "label1":"value1" "label2":"value2" "_msg":"foobar"`)
// Multiple streams
f(`{
@@ -113,7 +124,7 @@ func TestParseJSONRequest_Success(t *testing.T) {
]
}
]
}`, []int64{1577836800000000001, 1577836900005000002, 1877836900005000002}, `{"foo":"bar","a":"b","_msg":"foo bar"}
{"foo":"bar","a":"b","_msg":"abc"}
{"x":"y","_msg":"yx"}`)
}`, `_time:1577836800000000001 "foo":"bar" "a":"b" "_msg":"foo bar"
_time:1577836900005000002 "foo":"bar" "a":"b" "_msg":"abc"
_time:1877836900005000002 "x":"y" "_msg":"yx"`)
}

View File

@@ -6,7 +6,7 @@ import (
"testing"
"time"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
)
func BenchmarkParseJSONRequest(b *testing.B) {
@@ -22,13 +22,12 @@ func BenchmarkParseJSONRequest(b *testing.B) {
}
func benchmarkParseJSONRequest(b *testing.B, streams, rows, labels int) {
blp := &insertutils.BenchmarkLogMessageProcessor{}
b.ReportAllocs()
b.SetBytes(int64(streams * rows))
b.RunParallel(func(pb *testing.PB) {
data := getJSONBody(streams, rows, labels)
for pb.Next() {
_, err := parseJSONRequest(data, blp)
_, err := parseJSONRequest(data, func(_ int64, _ []logstorage.Field) {})
if err != nil {
panic(fmt.Errorf("unexpected error: %w", err))
}

View File

@@ -9,7 +9,6 @@ import (
"sync"
"time"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
@@ -44,9 +43,11 @@ func handleProtobuf(r *http.Request, w http.ResponseWriter) {
httpserver.Errorf(w, r, "%s", err)
return
}
lmp := cp.NewLogMessageProcessor()
n, err := parseProtobufRequest(data, lmp)
lmp.MustClose()
lr := logstorage.GetLogRows(cp.StreamFields, cp.IgnoreFields)
processLogMessage := cp.GetProcessLogMessageFunc(lr)
n, err := parseProtobufRequest(data, processLogMessage)
vlstorage.MustAddRows(lr)
logstorage.PutLogRows(lr)
if err != nil {
httpserver.Errorf(w, r, "cannot parse Loki protobuf request: %s", err)
return
@@ -66,7 +67,7 @@ var (
requestProtobufDuration = metrics.NewHistogram(`vl_http_request_duration_seconds{path="/insert/loki/api/v1/push",format="protobuf"}`)
)
func parseProtobufRequest(data []byte, lmp insertutils.LogMessageProcessor) (int, error) {
func parseProtobufRequest(data []byte, processLogMessage func(timestamp int64, fields []logstorage.Field)) (int, error) {
bb := bytesBufPool.Get()
defer bytesBufPool.Put(bb)
@@ -79,14 +80,12 @@ func parseProtobufRequest(data []byte, lmp insertutils.LogMessageProcessor) (int
req := getPushRequest()
defer putPushRequest(req)
err = req.UnmarshalProtobuf(bb.B)
err = req.Unmarshal(bb.B)
if err != nil {
return 0, fmt.Errorf("cannot parse request body: %w", err)
}
fields := getFields()
defer putFields(fields)
var commonFields []logstorage.Field
rowsIngested := 0
streams := req.Streams
currentTimestamp := time.Now().UnixNano()
@@ -94,60 +93,30 @@ func parseProtobufRequest(data []byte, lmp insertutils.LogMessageProcessor) (int
stream := &streams[i]
// st.Labels contains labels for the stream.
// Labels are same for all entries in the stream.
fields.fields, err = parsePromLabels(fields.fields[:0], stream.Labels)
commonFields, err = parsePromLabels(commonFields[:0], stream.Labels)
if err != nil {
return rowsIngested, fmt.Errorf("cannot parse stream labels %q: %w", stream.Labels, err)
}
commonFieldsLen := len(fields.fields)
fields := commonFields
entries := stream.Entries
for j := range entries {
e := &entries[j]
fields.fields = fields.fields[:commonFieldsLen]
for _, lp := range e.StructuredMetadata {
fields.fields = append(fields.fields, logstorage.Field{
Name: lp.Name,
Value: lp.Value,
})
}
fields.fields = append(fields.fields, logstorage.Field{
entry := &entries[j]
fields = append(fields[:len(commonFields)], logstorage.Field{
Name: "_msg",
Value: e.Line,
Value: entry.Line,
})
ts := e.Timestamp.UnixNano()
ts := entry.Timestamp.UnixNano()
if ts == 0 {
ts = currentTimestamp
}
lmp.AddRow(ts, fields.fields)
processLogMessage(ts, fields)
}
rowsIngested += len(stream.Entries)
}
return rowsIngested, nil
}
func getFields() *fields {
v := fieldsPool.Get()
if v == nil {
return &fields{}
}
return v.(*fields)
}
func putFields(f *fields) {
f.fields = f.fields[:0]
fieldsPool.Put(f)
}
var fieldsPool sync.Pool
type fields struct {
fields []logstorage.Field
}
// parsePromLabels parses log fields in Prometheus text exposition format from s, appends them to dst and returns the result.
//
// See test data of promtail for examples: https://github.com/grafana/loki/blob/a24ef7b206e0ca63ee74ca6ecb0a09b745cd2258/pkg/push/types_test.go
@@ -213,6 +182,6 @@ func getPushRequest() *PushRequest {
}
func putPushRequest(req *PushRequest) {
req.reset()
req.Reset()
pushReqsPool.Put(req)
}

View File

@@ -6,80 +6,83 @@ import (
"testing"
"time"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
"github.com/golang/snappy"
)
type testLogMessageProcessor struct {
pr PushRequest
}
func (tlp *testLogMessageProcessor) AddRow(timestamp int64, fields []logstorage.Field) {
msg := ""
for _, f := range fields {
if f.Name == "_msg" {
msg = f.Value
}
}
var a []string
for _, f := range fields {
if f.Name == "_msg" {
continue
}
item := fmt.Sprintf("%s=%q", f.Name, f.Value)
a = append(a, item)
}
labels := "{" + strings.Join(a, ", ") + "}"
tlp.pr.Streams = append(tlp.pr.Streams, Stream{
Labels: labels,
Entries: []Entry{
{
Timestamp: time.Unix(0, timestamp),
Line: strings.Clone(msg),
},
},
})
}
func (tlp *testLogMessageProcessor) MustClose() {
}
func TestParseProtobufRequest_Success(t *testing.T) {
f := func(s string, timestampsExpected []int64, resultExpected string) {
func TestParseProtobufRequestSuccess(t *testing.T) {
f := func(s string, resultExpected string) {
t.Helper()
tlp := &testLogMessageProcessor{}
n, err := parseJSONRequest([]byte(s), tlp)
var pr PushRequest
n, err := parseJSONRequest([]byte(s), func(timestamp int64, fields []logstorage.Field) {
msg := ""
for _, f := range fields {
if f.Name == "_msg" {
msg = f.Value
}
}
var a []string
for _, f := range fields {
if f.Name == "_msg" {
continue
}
item := fmt.Sprintf("%s=%q", f.Name, f.Value)
a = append(a, item)
}
labels := "{" + strings.Join(a, ", ") + "}"
pr.Streams = append(pr.Streams, Stream{
Labels: labels,
Entries: []Entry{
{
Timestamp: time.Unix(0, timestamp),
Line: msg,
},
},
})
})
if err != nil {
t.Fatalf("unexpected error: %s", err)
}
if n != len(tlp.pr.Streams) {
t.Fatalf("unexpected number of streams; got %d; want %d", len(tlp.pr.Streams), n)
if n != len(pr.Streams) {
t.Fatalf("unexpected number of streams; got %d; want %d", len(pr.Streams), n)
}
data := tlp.pr.MarshalProtobuf(nil)
data, err := pr.Marshal()
if err != nil {
t.Fatalf("unexpected error when marshaling PushRequest: %s", err)
}
encodedData := snappy.Encode(nil, data)
tlp2 := &insertutils.TestLogMessageProcessor{}
n, err = parseProtobufRequest(encodedData, tlp2)
var lines []string
n, err = parseProtobufRequest(encodedData, func(timestamp int64, fields []logstorage.Field) {
var a []string
for _, f := range fields {
a = append(a, f.String())
}
line := fmt.Sprintf("_time:%d %s", timestamp, strings.Join(a, " "))
lines = append(lines, line)
})
if err != nil {
t.Fatalf("unexpected error: %s", err)
}
if err := tlp2.Verify(n, timestampsExpected, resultExpected); err != nil {
t.Fatal(err)
if n != len(lines) {
t.Fatalf("unexpected number of lines parsed; got %d; want %d", n, len(lines))
}
result := strings.Join(lines, "\n")
if result != resultExpected {
t.Fatalf("unexpected result;\ngot\n%s\nwant\n%s", result, resultExpected)
}
}
// Empty streams
f(`{"streams":[]}`, nil, ``)
f(`{"streams":[{"values":[]}]}`, nil, ``)
f(`{"streams":[{"stream":{},"values":[]}]}`, nil, ``)
f(`{"streams":[{"stream":{"foo":"bar"},"values":[]}]}`, nil, ``)
f(`{"streams":[]}`, ``)
f(`{"streams":[{"values":[]}]}`, ``)
f(`{"streams":[{"stream":{},"values":[]}]}`, ``)
f(`{"streams":[{"stream":{"foo":"bar"},"values":[]}]}`, ``)
// Empty stream labels
f(`{"streams":[{"values":[["1577836800000000001", "foo bar"]]}]}`, []int64{1577836800000000001}, `{"_msg":"foo bar"}`)
f(`{"streams":[{"stream":{},"values":[["1577836800000000001", "foo bar"]]}]}`, []int64{1577836800000000001}, `{"_msg":"foo bar"}`)
f(`{"streams":[{"values":[["1577836800000000001", "foo bar"]]}]}`, `_time:1577836800000000001 "_msg":"foo bar"`)
f(`{"streams":[{"stream":{},"values":[["1577836800000000001", "foo bar"]]}]}`, `_time:1577836800000000001 "_msg":"foo bar"`)
// Non-empty stream labels
f(`{"streams":[{"stream":{
@@ -89,9 +92,9 @@ func TestParseProtobufRequest_Success(t *testing.T) {
["1577836800000000001", "foo bar"],
["1477836900005000002", "abc"],
["147.78369e9", "foobar"]
]}]}`, []int64{1577836800000000001, 1477836900005000002, 147783690000}, `{"label1":"value1","label2":"value2","_msg":"foo bar"}
{"label1":"value1","label2":"value2","_msg":"abc"}
{"label1":"value1","label2":"value2","_msg":"foobar"}`)
]}]}`, `_time:1577836800000000001 "label1":"value1" "label2":"value2" "_msg":"foo bar"
_time:1477836900005000002 "label1":"value1" "label2":"value2" "_msg":"abc"
_time:147783690000 "label1":"value1" "label2":"value2" "_msg":"foobar"`)
// Multiple streams
f(`{
@@ -115,12 +118,12 @@ func TestParseProtobufRequest_Success(t *testing.T) {
]
}
]
}`, []int64{1577836800000000001, 1577836900005000002, 1877836900005000002}, `{"foo":"bar","a":"b","_msg":"foo bar"}
{"foo":"bar","a":"b","_msg":"abc"}
{"x":"y","_msg":"yx"}`)
}`, `_time:1577836800000000001 "foo":"bar" "a":"b" "_msg":"foo bar"
_time:1577836900005000002 "foo":"bar" "a":"b" "_msg":"abc"
_time:1877836900005000002 "x":"y" "_msg":"yx"`)
}
func TestParsePromLabels_Success(t *testing.T) {
func TestParsePromLabelsSuccess(t *testing.T) {
f := func(s string) {
t.Helper()
fields, err := parsePromLabels(nil, s)
@@ -144,7 +147,7 @@ func TestParsePromLabels_Success(t *testing.T) {
f(`{foo="ba\"r\\z\n", a="", b="\"\\"}`)
}
func TestParsePromLabels_Failure(t *testing.T) {
func TestParsePromLabelsFailure(t *testing.T) {
f := func(s string) {
t.Helper()
fields, err := parsePromLabels(nil, s)

View File

@@ -8,8 +8,7 @@ import (
"github.com/golang/snappy"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
)
func BenchmarkParseProtobufRequest(b *testing.B) {
@@ -25,13 +24,12 @@ func BenchmarkParseProtobufRequest(b *testing.B) {
}
func benchmarkParseProtobufRequest(b *testing.B, streams, rows, labels int) {
blp := &insertutils.BenchmarkLogMessageProcessor{}
b.ReportAllocs()
b.SetBytes(int64(streams * rows))
b.RunParallel(func(pb *testing.PB) {
body := getProtobufBody(streams, rows, labels)
for pb.Next() {
_, err := parseProtobufRequest(body, blp)
_, err := parseProtobufRequest(body, func(_ int64, _ []logstorage.Field) {})
if err != nil {
panic(fmt.Errorf("unexpected error: %w", err))
}
@@ -39,47 +37,29 @@ func benchmarkParseProtobufRequest(b *testing.B, streams, rows, labels int) {
})
}
func getProtobufBody(streamsCount, rowsCount, labelsCount int) []byte {
var b []byte
var entries []Entry
streams := make([]Stream, streamsCount)
for i := range streams {
b = b[:0]
b = append(b, '{')
for j := 0; j < labelsCount; j++ {
b = append(b, "label_"...)
b = strconv.AppendInt(b, int64(j), 10)
b = append(b, `="value_`...)
b = strconv.AppendInt(b, int64(j), 10)
b = append(b, '"')
if j < labelsCount-1 {
b = append(b, ',')
func getProtobufBody(streams, rows, labels int) []byte {
var pr PushRequest
for i := 0; i < streams; i++ {
var st Stream
st.Labels = `{`
for j := 0; j < labels; j++ {
st.Labels += `label_` + strconv.Itoa(j) + `="value_` + strconv.Itoa(j) + `"`
if j < labels-1 {
st.Labels += `,`
}
}
b = append(b, '}')
labels := string(b)
st.Labels += `}`
var rowsBuf []byte
entriesLen := len(entries)
for j := 0; j < rowsCount; j++ {
rowsBufLen := len(rowsBuf)
rowsBuf = append(rowsBuf, "value_"...)
rowsBuf = strconv.AppendInt(rowsBuf, int64(j), 10)
entries = append(entries, Entry{
Timestamp: time.Now(),
Line: bytesutil.ToUnsafeString(rowsBuf[rowsBufLen:]),
})
for j := 0; j < rows; j++ {
st.Entries = append(st.Entries, Entry{Timestamp: time.Now(), Line: "value_" + strconv.Itoa(j)})
}
st := &streams[i]
st.Labels = labels
st.Entries = entries[entriesLen:]
}
pr := PushRequest{
Streams: streams,
pr.Streams = append(pr.Streams, st)
}
body := pr.MarshalProtobuf(nil)
body, _ := pr.Marshal()
encodedBody := snappy.Encode(nil, body)
return encodedBody

View File

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

File diff suppressed because it is too large Load Diff

View File

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

View File

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

481
app/vlinsert/loki/types.go Normal file
View File

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

View File

@@ -8,7 +8,6 @@ import (
"fmt"
"io"
"net"
"strconv"
"strings"
"sync"
"sync/atomic"
@@ -26,49 +25,34 @@ import (
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/netutil"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/slicesutil"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/writeconcurrencylimiter"
"github.com/VictoriaMetrics/metrics"
)
var (
syslogTenantID = flag.String("syslog.tenantID", "0:0", "TenantID for logs ingested via Syslog protocol. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
syslogTimezone = flag.String("syslog.timezone", "Local", "Timezone to use when parsing timestamps in RFC3164 syslog messages. Timezone must be a valid IANA Time Zone. "+
"For example: America/New_York, Europe/Berlin, Etc/GMT+3 . See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
syslogTenantIDTCP = flagutil.NewArrayString("syslog.tenantID.tcp", "TenantID for logs ingested via the corresponding -syslog.listenAddr.tcp. "+
"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
syslogTenantIDUDP = flagutil.NewArrayString("syslog.tenantID.udp", "TenantID for logs ingested via the corresponding -syslog.listenAddr.udp. "+
"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
listenAddrTCP = flag.String("syslog.listenAddr.tcp", "", "Optional TCP address to listen to for Syslog messages. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
listenAddrUDP = flag.String("syslog.listenAddr.udp", "", "Optional UDP address to listen to for Syslog messages. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
listenAddrTCP = flagutil.NewArrayString("syslog.listenAddr.tcp", "Comma-separated list of TCP addresses to listen to for Syslog messages. "+
"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
listenAddrUDP = flagutil.NewArrayString("syslog.listenAddr.udp", "Comma-separated list of UDP address to listen to for Syslog messages. "+
"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
tlsEnable = flagutil.NewArrayBool("syslog.tls", "Whether to enable TLS for receiving syslog messages at the corresponding -syslog.listenAddr.tcp. "+
"The corresponding -syslog.tlsCertFile and -syslog.tlsKeyFile must be set if -syslog.tls is set. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#security")
tlsCertFile = flagutil.NewArrayString("syslog.tlsCertFile", "Path to file with TLS certificate for the corresponding -syslog.listenAddr.tcp if the corresponding -syslog.tls is set. "+
tlsEnable = flag.Bool("syslog.tls", false, "Whether to use TLS for receiving syslog messages at -syslog.listenAddr.tcp. -syslog.tlsCertFile and -syslog.tlsKeyFile must be set "+
"if -syslog.tls is set. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
tlsCertFile = flag.String("syslog.tlsCertFile", "", "Path to file with TLS certificate for -syslog.listenAddr.tcp if -syslog.tls is set. "+
"Prefer ECDSA certs instead of RSA certs as RSA certs are slower. The provided certificate file is automatically re-read every second, so it can be dynamically updated. "+
"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#security")
tlsKeyFile = flagutil.NewArrayString("syslog.tlsKeyFile", "Path to file with TLS key for the corresponding -syslog.listenAddr.tcp if the corresponding -syslog.tls is set. "+
"The provided key file is automatically re-read every second, so it can be dynamically updated. "+
"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#security")
"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
tlsKeyFile = flag.String("syslog.tlsKeyFile", "", "Path to file with TLS key for -syslog.listenAddr.tcp if -syslog.tls is set. "+
"The provided key file is automatically re-read every second, so it can be dynamically updated")
tlsCipherSuites = flagutil.NewArrayString("syslog.tlsCipherSuites", "Optional list of TLS cipher suites for -syslog.listenAddr.tcp if -syslog.tls is set. "+
"See the list of supported cipher suites at https://pkg.go.dev/crypto/tls#pkg-constants . "+
"See also https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#security")
"See also https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
tlsMinVersion = flag.String("syslog.tlsMinVersion", "TLS13", "The minimum TLS version to use for -syslog.listenAddr.tcp if -syslog.tls is set. "+
"Supported values: TLS10, TLS11, TLS12, TLS13. "+
"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#security")
"See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
compressMethodTCP = flagutil.NewArrayString("syslog.compressMethod.tcp", "Compression method for syslog messages received at the corresponding -syslog.listenAddr.tcp. "+
"Supported values: none, gzip, deflate. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#compression")
compressMethodUDP = flagutil.NewArrayString("syslog.compressMethod.udp", "Compression method for syslog messages received at the corresponding -syslog.listenAddr.udp. "+
"Supported values: none, gzip, deflate. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#compression")
useLocalTimestampTCP = flagutil.NewArrayBool("syslog.useLocalTimestamp.tcp", "Whether to use local timestamp instead of the original timestamp for the ingested syslog messages "+
"at the corresponding -syslog.listenAddr.tcp. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#log-timestamps")
useLocalTimestampUDP = flagutil.NewArrayBool("syslog.useLocalTimestamp.udp", "Whether to use local timestamp instead of the original timestamp for the ingested syslog messages "+
"at the corresponding -syslog.listenAddr.udp. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#log-timestamps")
compressMethod = flag.String("syslog.compressMethod", "", "Compression method for syslog messages received at -syslog.listenAddr.tcp and -syslog.listenAddr.udp. "+
"Supported values: none, gzip, deflate. See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/")
)
// MustInit initializes syslog parser at the given -syslog.listenAddr.tcp and -syslog.listenAddr.udp ports
@@ -82,20 +66,32 @@ func MustInit() {
}
workersStopCh = make(chan struct{})
for argIdx, addr := range *listenAddrTCP {
workersWG.Add(1)
go func(addr string, argIdx int) {
runTCPListener(addr, argIdx)
workersWG.Done()
}(addr, argIdx)
tenantID, err := logstorage.GetTenantIDFromString(*syslogTenantID)
if err != nil {
logger.Fatalf("cannot parse -syslog.tenantID=%q: %s", *syslogTenantID, err)
}
globalTenantID = tenantID
switch *compressMethod {
case "", "none", "gzip", "deflate":
default:
logger.Fatalf("unexpected -syslog.compressLevel=%q; supported values: none, gzip, deflate", *compressMethod)
}
for argIdx, addr := range *listenAddrUDP {
if *listenAddrTCP != "" {
workersWG.Add(1)
go func(addr string, argIdx int) {
runUDPListener(addr, argIdx)
go func() {
runTCPListener(*listenAddrTCP)
workersWG.Done()
}(addr, argIdx)
}()
}
if *listenAddrUDP != "" {
workersWG.Add(1)
go func() {
runUDPListener(*listenAddrUDP)
workersWG.Done()
}()
}
currentYear := time.Now().Year()
@@ -128,6 +124,7 @@ func MustInit() {
}
var (
globalTenantID logstorage.TenantID
globalCurrentYear atomic.Int64
globalTimezone *time.Location
)
@@ -144,47 +141,32 @@ func MustStop() {
workersStopCh = nil
}
func runUDPListener(addr string, argIdx int) {
func runUDPListener(addr string) {
ln, err := net.ListenPacket(netutil.GetUDPNetwork(), addr)
if err != nil {
logger.Fatalf("cannot start UDP syslog server at %q: %s", addr, err)
}
tenantIDStr := syslogTenantIDUDP.GetOptionalArg(argIdx)
tenantID, err := logstorage.ParseTenantID(tenantIDStr)
if err != nil {
logger.Fatalf("cannot parse -syslog.tenantID.udp=%q for -syslog.listenAddr.udp=%q: %s", tenantIDStr, addr, err)
}
compressMethod := compressMethodUDP.GetOptionalArg(argIdx)
checkCompressMethod(compressMethod, addr, "udp")
useLocalTimestamp := useLocalTimestampUDP.GetOptionalArg(argIdx)
doneCh := make(chan struct{})
go func() {
serveUDP(ln, tenantID, compressMethod, useLocalTimestamp)
serveUDP(ln)
close(doneCh)
}()
logger.Infof("started accepting syslog messages at -syslog.listenAddr.udp=%q", addr)
<-workersStopCh
if err := ln.Close(); err != nil {
logger.Fatalf("syslog: cannot close UDP listener at %s: %s", addr, err)
}
<-doneCh
logger.Infof("finished accepting syslog messages at -syslog.listenAddr.udp=%q", addr)
}
func runTCPListener(addr string, argIdx int) {
func runTCPListener(addr string) {
var tlsConfig *tls.Config
if tlsEnable.GetOptionalArg(argIdx) {
certFile := tlsCertFile.GetOptionalArg(argIdx)
keyFile := tlsKeyFile.GetOptionalArg(argIdx)
tc, err := netutil.GetServerTLSConfig(certFile, keyFile, *tlsMinVersion, *tlsCipherSuites)
if *tlsEnable {
tc, err := netutil.GetServerTLSConfig(*tlsCertFile, *tlsKeyFile, *tlsMinVersion, *tlsCipherSuites)
if err != nil {
logger.Fatalf("cannot load TLS cert from -syslog.tlsCertFile=%q, -syslog.tlsKeyFile=%q, -syslog.tlsMinVersion=%q, -syslog.tlsCipherSuites=%q: %s",
certFile, keyFile, *tlsMinVersion, *tlsCipherSuites, err)
*tlsCertFile, *tlsKeyFile, *tlsMinVersion, *tlsCipherSuites, err)
}
tlsConfig = tc
}
@@ -193,42 +175,20 @@ func runTCPListener(addr string, argIdx int) {
logger.Fatalf("syslog: cannot start TCP listener at %s: %s", addr, err)
}
tenantIDStr := syslogTenantIDTCP.GetOptionalArg(argIdx)
tenantID, err := logstorage.ParseTenantID(tenantIDStr)
if err != nil {
logger.Fatalf("cannot parse -syslog.tenantID.tcp=%q for -syslog.listenAddr.tcp=%q: %s", tenantIDStr, addr, err)
}
compressMethod := compressMethodTCP.GetOptionalArg(argIdx)
checkCompressMethod(compressMethod, addr, "tcp")
useLocalTimestamp := useLocalTimestampTCP.GetOptionalArg(argIdx)
doneCh := make(chan struct{})
go func() {
serveTCP(ln, tenantID, compressMethod, useLocalTimestamp)
serveTCP(ln)
close(doneCh)
}()
logger.Infof("started accepting syslog messages at -syslog.listenAddr.tcp=%q", addr)
<-workersStopCh
if err := ln.Close(); err != nil {
logger.Fatalf("syslog: cannot close TCP listener at %s: %s", addr, err)
}
<-doneCh
logger.Infof("finished accepting syslog messages at -syslog.listenAddr.tcp=%q", addr)
}
func checkCompressMethod(compressMethod, addr, protocol string) {
switch compressMethod {
case "", "none", "gzip", "deflate":
return
default:
logger.Fatalf("unsupported -syslog.compressMethod.%s=%q for -syslog.listenAddr.%s=%q; supported values: 'none', 'gzip', 'deflate'", protocol, compressMethod, protocol, addr)
}
}
func serveUDP(ln net.PacketConn, tenantID logstorage.TenantID, compressMethod string, useLocalTimestamp bool) {
func serveUDP(ln net.PacketConn) {
gomaxprocs := cgroup.AvailableCPUs()
var wg sync.WaitGroup
localAddr := ln.LocalAddr()
@@ -236,7 +196,7 @@ func serveUDP(ln net.PacketConn, tenantID logstorage.TenantID, compressMethod st
wg.Add(1)
go func() {
defer wg.Done()
cp := insertutils.GetCommonParamsForSyslog(tenantID)
cp := insertutils.GetCommonParamsForSyslog(globalTenantID)
var bb bytesutil.ByteBuffer
bb.B = bytesutil.ResizeNoCopyNoOverallocate(bb.B, 64*1024)
for {
@@ -261,7 +221,7 @@ func serveUDP(ln net.PacketConn, tenantID logstorage.TenantID, compressMethod st
}
bb.B = bb.B[:n]
udpRequestsTotal.Inc()
if err := processStream(bb.NewReader(), compressMethod, useLocalTimestamp, cp); err != nil {
if err := processStream(bb.NewReader(), cp); err != nil {
logger.Errorf("syslog: cannot process UDP data from %s at %s: %s", remoteAddr, localAddr, err)
}
}
@@ -270,7 +230,7 @@ func serveUDP(ln net.PacketConn, tenantID logstorage.TenantID, compressMethod st
wg.Wait()
}
func serveTCP(ln net.Listener, tenantID logstorage.TenantID, compressMethod string, useLocalTimestamp bool) {
func serveTCP(ln net.Listener) {
var cm ingestserver.ConnsMap
cm.Init("syslog")
@@ -300,8 +260,8 @@ func serveTCP(ln net.Listener, tenantID logstorage.TenantID, compressMethod stri
wg.Add(1)
go func() {
cp := insertutils.GetCommonParamsForSyslog(tenantID)
if err := processStream(c, compressMethod, useLocalTimestamp, cp); err != nil {
cp := insertutils.GetCommonParamsForSyslog(globalTenantID)
if err := processStream(c, cp); err != nil {
logger.Errorf("syslog: cannot process TCP data at %q: %s", addr, err)
}
@@ -316,20 +276,8 @@ func serveTCP(ln net.Listener, tenantID logstorage.TenantID, compressMethod stri
}
// processStream parses a stream of syslog messages from r and ingests them into vlstorage.
func processStream(r io.Reader, compressMethod string, useLocalTimestamp bool, cp *insertutils.CommonParams) error {
if err := vlstorage.CanWriteData(); err != nil {
return err
}
lmp := cp.NewLogMessageProcessor()
err := processStreamInternal(r, compressMethod, useLocalTimestamp, lmp)
lmp.MustClose()
return err
}
func processStreamInternal(r io.Reader, compressMethod string, useLocalTimestamp bool, lmp insertutils.LogMessageProcessor) error {
switch compressMethod {
func processStream(r io.Reader, cp *insertutils.CommonParams) error {
switch *compressMethod {
case "", "none":
case "gzip":
zr, err := common.GetGzipReader(r)
@@ -344,12 +292,12 @@ func processStreamInternal(r io.Reader, compressMethod string, useLocalTimestamp
}
r = zr
default:
logger.Panicf("BUG: unsupported compressMethod=%q; supported values: none, gzip, deflate", compressMethod)
logger.Panicf("BUG: compressLevel=%q; supported values: none, gzip, deflate", *compressMethod)
}
err := processUncompressedStream(r, useLocalTimestamp, lmp)
err := processUncompressedStream(r, cp)
switch compressMethod {
switch *compressMethod {
case "gzip":
zr := r.(*gzip.Reader)
common.PutGzipReader(zr)
@@ -361,166 +309,76 @@ func processStreamInternal(r io.Reader, compressMethod string, useLocalTimestamp
return err
}
func processUncompressedStream(r io.Reader, useLocalTimestamp bool, lmp insertutils.LogMessageProcessor) error {
func processUncompressedStream(r io.Reader, cp *insertutils.CommonParams) error {
if err := vlstorage.CanWriteData(); err != nil {
return err
}
lr := logstorage.GetLogRows(cp.StreamFields, nil)
processLogMessage := cp.GetProcessLogMessageFunc(lr)
wcr := writeconcurrencylimiter.GetReader(r)
defer writeconcurrencylimiter.PutReader(wcr)
slr := getSyslogLineReader(wcr)
defer putSyslogLineReader(slr)
lb := lineBufferPool.Get()
defer lineBufferPool.Put(lb)
lb.B = bytesutil.ResizeNoCopyNoOverallocate(lb.B, insertutils.MaxLineSizeBytes.IntN())
sc := bufio.NewScanner(wcr)
sc.Buffer(lb.B, len(lb.B))
n := 0
for {
ok := slr.nextLine()
wcr.DecConcurrency()
if !ok {
break
}
currentYear := int(globalCurrentYear.Load())
err := processLine(slr.line, currentYear, globalTimezone, useLocalTimestamp, lmp)
ok, err := readLine(sc, currentYear, globalTimezone, processLogMessage)
wcr.DecConcurrency()
if err != nil {
errorsTotal.Inc()
return fmt.Errorf("cannot read line #%d: %s", n, err)
}
if !ok {
break
}
n++
rowsIngestedTotal.Inc()
}
return slr.Error()
}
type syslogLineReader struct {
line []byte
br *bufio.Reader
err error
}
func (slr *syslogLineReader) reset(r io.Reader) {
slr.line = slr.line[:0]
slr.br.Reset(r)
slr.err = nil
}
// Error returns the last error occurred in slr.
func (slr *syslogLineReader) Error() error {
if slr.err == nil || slr.err == io.EOF {
return nil
}
return slr.err
}
// nextLine reads the next syslog line from slr and stores it at slr.line.
//
// false is returned if the next line cannot be read. Error() must be called in this case
// in order to verify whether there is an error or just slr stream has been finished.
func (slr *syslogLineReader) nextLine() bool {
if slr.err != nil {
return false
}
again:
prefix, err := slr.br.ReadSlice(' ')
if err != nil {
if err != io.EOF {
slr.err = fmt.Errorf("cannot read message frame prefix: %w", err)
return false
}
if len(prefix) == 0 {
slr.err = err
return false
}
}
// skip empty lines
for len(prefix) > 0 && prefix[0] == '\n' {
prefix = prefix[1:]
}
if len(prefix) == 0 {
// An empty prefix or a prefix with empty lines - try reading yet another prefix.
goto again
}
if prefix[0] >= '0' && prefix[0] <= '9' {
// This is octet-counting method. See https://www.ietf.org/archive/id/draft-gerhards-syslog-plain-tcp-07.html#msgxfer
msgLenStr := bytesutil.ToUnsafeString(prefix[:len(prefix)-1])
msgLen, err := strconv.ParseUint(msgLenStr, 10, 64)
if err != nil {
slr.err = fmt.Errorf("cannot parse message length from %q: %w", msgLenStr, err)
return false
}
if maxMsgLen := insertutils.MaxLineSizeBytes.IntN(); msgLen > uint64(maxMsgLen) {
slr.err = fmt.Errorf("cannot read message longer than %d bytes; msgLen=%d", maxMsgLen, msgLen)
return false
}
slr.line = slicesutil.SetLength(slr.line, int(msgLen))
if _, err := io.ReadFull(slr.br, slr.line); err != nil {
slr.err = fmt.Errorf("cannot read message with size %d bytes: %w", msgLen, err)
return false
}
return true
}
// This is octet-stuffing method. See https://www.ietf.org/archive/id/draft-gerhards-syslog-plain-tcp-07.html#octet-stuffing-legacy
slr.line = append(slr.line[:0], prefix...)
for {
line, err := slr.br.ReadSlice('\n')
if err == nil {
slr.line = append(slr.line, line[:len(line)-1]...)
return true
}
if err == io.EOF {
slr.line = append(slr.line, line...)
return true
}
if err == bufio.ErrBufferFull {
slr.line = append(slr.line, line...)
continue
}
slr.err = fmt.Errorf("cannot read message in octet-stuffing method: %w", err)
return false
}
}
func getSyslogLineReader(r io.Reader) *syslogLineReader {
v := syslogLineReaderPool.Get()
if v == nil {
br := bufio.NewReaderSize(r, 64*1024)
return &syslogLineReader{
br: br,
}
}
slr := v.(*syslogLineReader)
slr.reset(r)
return slr
}
func putSyslogLineReader(slr *syslogLineReader) {
syslogLineReaderPool.Put(slr)
}
var syslogLineReaderPool sync.Pool
func processLine(line []byte, currentYear int, timezone *time.Location, useLocalTimestamp bool, lmp insertutils.LogMessageProcessor) error {
p := logstorage.GetSyslogParser(currentYear, timezone)
lineStr := bytesutil.ToUnsafeString(line)
p.Parse(lineStr)
var ts int64
if useLocalTimestamp {
ts = time.Now().UnixNano()
} else {
nsecs, err := insertutils.ExtractTimestampRFC3339NanoFromFields("timestamp", p.Fields)
if err != nil {
return fmt.Errorf("cannot get timestamp from syslog line %q: %w", line, err)
}
ts = nsecs
}
logstorage.RenameField(p.Fields, "message", "_msg")
lmp.AddRow(ts, p.Fields)
logstorage.PutSyslogParser(p)
vlstorage.MustAddRows(lr)
logstorage.PutLogRows(lr)
return nil
}
func readLine(sc *bufio.Scanner, currentYear int, timezone *time.Location, processLogMessage func(timestamp int64, fields []logstorage.Field)) (bool, error) {
var line []byte
for len(line) == 0 {
if !sc.Scan() {
if err := sc.Err(); err != nil {
if errors.Is(err, bufio.ErrTooLong) {
return false, fmt.Errorf(`line size exceeds -insert.maxLineSizeBytes=%d`, insertutils.MaxLineSizeBytes.IntN())
}
return false, err
}
return false, nil
}
line = sc.Bytes()
}
p := logstorage.GetSyslogParser(currentYear, timezone)
lineStr := bytesutil.ToUnsafeString(line)
p.Parse(lineStr)
ts, err := insertutils.ExtractTimestampISO8601FromFields("timestamp", p.Fields)
if err != nil {
return false, fmt.Errorf("cannot get timestamp from syslog line %q: %w", line, err)
}
logstorage.RenameField(p.Fields, "message", "_msg")
processLogMessage(ts, p.Fields)
logstorage.PutSyslogParser(p)
return true, nil
}
var lineBufferPool bytesutil.ByteBufferPool
var (
rowsIngestedTotal = metrics.NewCounter(`vl_rows_ingested_total{type="syslog"}`)

View File

@@ -1,130 +1,65 @@
package syslog
import (
"bufio"
"bytes"
"reflect"
"testing"
"time"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
)
func TestSyslogLineReader_Success(t *testing.T) {
f := func(data string, linesExpected []string) {
t.Helper()
r := bytes.NewBufferString(data)
slr := getSyslogLineReader(r)
defer putSyslogLineReader(slr)
var lines []string
for slr.nextLine() {
lines = append(lines, string(slr.line))
}
if err := slr.Error(); err != nil {
t.Fatalf("unexpected error: %s", err)
}
if !reflect.DeepEqual(lines, linesExpected) {
t.Fatalf("unexpected lines read;\ngot\n%q\nwant\n%q", lines, linesExpected)
}
}
f("", nil)
f("\n", nil)
f("\n\n\n", nil)
f("foobar", []string{"foobar"})
f("foobar\n", []string{"foobar\n"})
f("\n\nfoo\n\nbar\n\n", []string{"foo\n\nbar\n\n"})
f(`Jun 3 12:08:33 abcd systemd: Starting Update the local ESM caches...`, []string{"Jun 3 12:08:33 abcd systemd: Starting Update the local ESM caches..."})
f(`Jun 3 12:08:33 abcd systemd: Starting Update the local ESM caches...
48 <165>Jun 4 12:08:33 abcd systemd[345]: abc defg<123>1 2023-06-03T17:42:12.345Z mymachine.example.com appname 12345 ID47 [exampleSDID@32473 iut="3" eventSource="Application 123 = ] 56" eventID="11211"] This is a test message with structured data.
`, []string{
"Jun 3 12:08:33 abcd systemd: Starting Update the local ESM caches...",
"<165>Jun 4 12:08:33 abcd systemd[345]: abc defg",
`<123>1 2023-06-03T17:42:12.345Z mymachine.example.com appname 12345 ID47 [exampleSDID@32473 iut="3" eventSource="Application 123 = ] 56" eventID="11211"] This is a test message with structured data.`,
})
}
func TestSyslogLineReader_Failure(t *testing.T) {
f := func(data string) {
t.Helper()
r := bytes.NewBufferString(data)
slr := getSyslogLineReader(r)
defer putSyslogLineReader(slr)
if slr.nextLine() {
t.Fatalf("expecting failure to read the first line")
}
if err := slr.Error(); err == nil {
t.Fatalf("expecting non-nil error")
}
}
// invalid format for message size
f("12foo bar")
// too big message size
f("123 aa")
f("1233423432 abc")
}
func TestProcessStreamInternal_Success(t *testing.T) {
func TestReadLine_Success(t *testing.T) {
f := func(data string, currentYear, rowsExpected int, timestampsExpected []int64, resultExpected string) {
t.Helper()
MustInit()
defer MustStop()
globalTimezone = time.UTC
globalCurrentYear.Store(int64(currentYear))
tlp := &insertutils.TestLogMessageProcessor{}
r := bytes.NewBufferString(data)
if err := processStreamInternal(r, "", false, tlp); err != nil {
t.Fatalf("unexpected error: %s", err)
var timestamps []int64
var result string
processLogMessage := func(timestamp int64, fields []logstorage.Field) {
timestamps = append(timestamps, timestamp)
result += string(logstorage.MarshalFieldsToJSON(nil, fields)) + "\n"
}
if err := tlp.Verify(rowsExpected, timestampsExpected, resultExpected); err != nil {
t.Fatal(err)
r := bytes.NewBufferString(data)
sc := bufio.NewScanner(r)
rows := 0
timezone := time.UTC
for {
ok, err := readLine(sc, currentYear, timezone, processLogMessage)
if err != nil {
t.Fatalf("unexpected error: %s", err)
}
if !ok {
break
}
rows++
}
if rows != rowsExpected {
t.Fatalf("unexpected rows read; got %d; want %d", rows, rowsExpected)
}
if !reflect.DeepEqual(timestamps, timestampsExpected) {
t.Fatalf("unexpected timestamps;\ngot\n%d\nwant\n%d", timestamps, timestampsExpected)
}
if result != resultExpected {
t.Fatalf("unexpected result;\ngot\n%s\nwant\n%s", result, resultExpected)
}
}
data := `Jun 3 12:08:33 abcd systemd: Starting Update the local ESM caches...
48 <165>Jun 4 12:08:33 abcd systemd[345]: abc defg<123>1 2023-06-03T17:42:12.345Z mymachine.example.com appname 12345 ID47 [exampleSDID@32473 iut="3" eventSource="Application 123 = ] 56" eventID="11211"] This is a test message with structured data.
<165>Jun 4 12:08:33 abcd systemd[345]: abc defg
<123>1 2023-06-03T17:42:12.345Z mymachine.example.com appname 12345 ID47 [exampleSDID@32473 iut="3" eventSource="Application 123 = ] 56" eventID="11211"] This is a test message with structured data.
`
currentYear := 2023
rowsExpected := 3
timestampsExpected := []int64{1685794113000000000, 1685880513000000000, 1685814132345000000}
resultExpected := `{"format":"rfc3164","timestamp":"","hostname":"abcd","app_name":"systemd","_msg":"Starting Update the local ESM caches..."}
{"priority":"165","facility":"20","severity":"5","format":"rfc3164","timestamp":"","hostname":"abcd","app_name":"systemd","proc_id":"345","_msg":"abc defg"}
{"priority":"123","facility":"15","severity":"3","format":"rfc5424","timestamp":"","hostname":"mymachine.example.com","app_name":"appname","proc_id":"12345","msg_id":"ID47","exampleSDID@32473.iut":"3","exampleSDID@32473.eventSource":"Application 123 = ] 56","exampleSDID@32473.eventID":"11211","_msg":"This is a test message with structured data."}`
{"priority":"123","facility":"15","severity":"3","format":"rfc5424","timestamp":"","hostname":"mymachine.example.com","app_name":"appname","proc_id":"12345","msg_id":"ID47","exampleSDID@32473":"iut=\"3\" eventSource=\"Application 123 = ] 56\" eventID=\"11211\"","_msg":"This is a test message with structured data."}
`
f(data, currentYear, rowsExpected, timestampsExpected, resultExpected)
}
func TestProcessStreamInternal_Failure(t *testing.T) {
f := func(data string) {
t.Helper()
MustInit()
defer MustStop()
tlp := &insertutils.TestLogMessageProcessor{}
r := bytes.NewBufferString(data)
if err := processStreamInternal(r, "", false, tlp); err == nil {
t.Fatalf("expecting non-nil error")
}
}
// invalid format for message size
f("12foo bar")
// too big message size
f("123 foo")
f("123456789 bar")
}

View File

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

View File

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

View File

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

View File

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

View File

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

View File

@@ -1,13 +1,13 @@
{
"files": {
"main.css": "./static/css/main.1041c3d4.css",
"main.js": "./static/js/main.8451d9b9.js",
"main.css": "./static/css/main.2fa7c03f.css",
"main.js": "./static/js/main.68f1bd69.js",
"static/js/685.bebe1265.chunk.js": "./static/js/685.bebe1265.chunk.js",
"static/media/MetricsQL.md": "./static/media/MetricsQL.d46c42c8e891f06298c4.md",
"static/media/MetricsQL.md": "./static/media/MetricsQL.cb83d071da309a358bc0.md",
"index.html": "./index.html"
},
"entrypoints": [
"static/css/main.1041c3d4.css",
"static/js/main.8451d9b9.js"
"static/css/main.2fa7c03f.css",
"static/js/main.68f1bd69.js"
]
}

View File

@@ -1 +1 @@
<!doctype html><html lang="en"><head><meta charset="utf-8"/><link rel="icon" href="./favicon.ico"/><meta name="viewport" content="width=device-width,initial-scale=1,maximum-scale=5"/><meta name="theme-color" content="#000000"/><meta name="description" content="UI for VictoriaMetrics"/><link rel="apple-touch-icon" href="./apple-touch-icon.png"/><link rel="icon" type="image/png" sizes="32x32" href="./favicon-32x32.png"><link rel="manifest" href="./manifest.json"/><title>VM UI</title><script src="./dashboards/index.js" type="module"></script><meta name="twitter:card" content="summary_large_image"><meta name="twitter:image" content="./preview.jpg"><meta name="twitter:title" content="UI for VictoriaMetrics"><meta name="twitter:description" content="Explore and troubleshoot your VictoriaMetrics data"><meta name="twitter:site" content="@VictoriaMetrics"><meta property="og:title" content="Metric explorer for VictoriaMetrics"><meta property="og:description" content="Explore and troubleshoot your VictoriaMetrics data"><meta property="og:image" content="./preview.jpg"><meta property="og:type" content="website"><script defer="defer" src="./static/js/main.8451d9b9.js"></script><link href="./static/css/main.1041c3d4.css" rel="stylesheet"></head><body><noscript>You need to enable JavaScript to run this app.</noscript><div id="root"></div></body></html>
<!doctype html><html lang="en"><head><meta charset="utf-8"/><link rel="icon" href="./favicon.ico"/><meta name="viewport" content="width=device-width,initial-scale=1,maximum-scale=5"/><meta name="theme-color" content="#000000"/><meta name="description" content="UI for VictoriaMetrics"/><link rel="apple-touch-icon" href="./apple-touch-icon.png"/><link rel="icon" type="image/png" sizes="32x32" href="./favicon-32x32.png"><link rel="manifest" href="./manifest.json"/><title>VM UI</title><script src="./dashboards/index.js" type="module"></script><meta name="twitter:card" content="summary_large_image"><meta name="twitter:image" content="./preview.jpg"><meta name="twitter:title" content="UI for VictoriaMetrics"><meta name="twitter:description" content="Explore and troubleshoot your VictoriaMetrics data"><meta name="twitter:site" content="@VictoriaMetrics"><meta property="og:title" content="Metric explorer for VictoriaMetrics"><meta property="og:description" content="Explore and troubleshoot your VictoriaMetrics data"><meta property="og:image" content="./preview.jpg"><meta property="og:type" content="website"><script defer="defer" src="./static/js/main.68f1bd69.js"></script><link href="./static/css/main.2fa7c03f.css" rel="stylesheet"></head><body><noscript>You need to enable JavaScript to run this app.</noscript><div id="root"></div></body></html>

File diff suppressed because one or more lines are too long

File diff suppressed because one or more lines are too long

File diff suppressed because one or more lines are too long

View File

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

File diff suppressed because one or more lines are too long

View File

@@ -10,6 +10,9 @@ aliases:
- /ExtendedPromQL.html
- /MetricsQL.html
---
# MetricsQL
[VictoriaMetrics](https://github.com/VictoriaMetrics/VictoriaMetrics) implements MetricsQL -
query language inspired by [PromQL](https://prometheus.io/docs/prometheus/latest/querying/basics/).
MetricsQL is backwards-compatible with PromQL, so Grafana dashboards backed by Prometheus datasource should work
@@ -104,7 +107,7 @@ The list of MetricsQL features on top of PromQL:
* Trailing commas on all the lists are allowed - label filters, function args and with expressions.
For instance, the following queries are valid: `m{foo="bar",}`, `f(a, b,)`, `WITH (x=y,) x`.
This simplifies maintenance of multi-line queries.
* Metric names and label names may contain any unicode letter. For example `температура{город="Київ"}` is a valid MetricsQL expression.
* Metric names and label names may contain any unicode letter. For example `температура{город="Київ"}` is a value MetricsQL expression.
* Metric names and labels names may contain escaped chars. For example, `foo\-bar{baz\=aa="b"}` is valid expression.
It returns time series with name `foo-bar` containing label `baz=aa` with value `b`.
Additionally, the following escape sequences are supported:
@@ -217,11 +220,9 @@ See also [descent_over_time](#descent_over_time).
over [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples) on the given lookbehind window `d` per each time series returned
from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
This function is supported by PromQL.
See also [median_over_time](#median_over_time), [min_over_time](#min_over_time) and [max_over_time](#max_over_time).
See also [median_over_time](#median_over_time).
#### changes
@@ -261,8 +262,6 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [count_over_time](#count_over_time), [share_eq_over_time](#share_eq_over_time) and [count_values_over_time](#count_values_over_time).
#### count_gt_over_time
@@ -273,8 +272,6 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [count_over_time](#count_over_time) and [share_gt_over_time](#share_gt_over_time).
#### count_le_over_time
@@ -285,8 +282,6 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [count_over_time](#count_over_time) and [share_le_over_time](#share_le_over_time).
#### count_ne_over_time
@@ -297,9 +292,7 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [count_over_time](#count_over_time) and [count_eq_over_time](#count_eq_over_time).
See also [count_over_time](#count_over_time).
#### count_over_time
@@ -320,8 +313,6 @@ The results are calculated independently per each time series returned from the
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [count_eq_over_time](#count_eq_over_time), [count_values](#count_values) and [distinct_over_time](#distinct_over_time) and [label_match](#label_match).
#### decreases_over_time
@@ -441,8 +432,6 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
#### histogram_over_time
`histogram_over_time(series_selector[d])` is a [rollup function](#rollup-functions), which calculates
@@ -454,15 +443,11 @@ For example, the following query calculates median temperature by country over t
`histogram_quantile(0.5, sum(histogram_over_time(temperature[24h])) by (vmrange,country))`.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
#### hoeffding_bound_lower
`hoeffding_bound_lower(phi, series_selector[d])` is a [rollup function](#rollup-functions), which calculates
lower [Hoeffding bound](https://en.wikipedia.org/wiki/Hoeffding%27s_inequality) for the given `phi` in the range `[0...1]`.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [hoeffding_bound_upper](#hoeffding_bound_upper).
#### hoeffding_bound_upper
@@ -470,8 +455,6 @@ See also [hoeffding_bound_upper](#hoeffding_bound_upper).
`hoeffding_bound_upper(phi, series_selector[d])` is a [rollup function](#rollup-functions), which calculates
upper [Hoeffding bound](https://en.wikipedia.org/wiki/Hoeffding%27s_inequality) for the given `phi` in the range `[0...1]`.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [hoeffding_bound_lower](#hoeffding_bound_lower).
#### holt_winters
@@ -479,9 +462,8 @@ See also [hoeffding_bound_lower](#hoeffding_bound_lower).
`holt_winters(series_selector[d], sf, tf)` is a [rollup function](#rollup-functions), which calculates Holt-Winters value
(aka [double exponential smoothing](https://en.wikipedia.org/wiki/Exponential_smoothing#Double_exponential_smoothing)) for [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
over the given lookbehind window `d` using the given smoothing factor `sf` and the given trend factor `tf`.
Both `sf` and `tf` must be in the range `[0...1]`.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
Both `sf` and `tf` must be in the range `[0...1]`. It is expected that the [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering)
returns time series of [gauge type](https://docs.victoriametrics.com/keyconcepts/#gauge).
This function is supported by PromQL.
@@ -513,14 +495,13 @@ See also [deriv](#deriv).
`increase(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the increase over the given lookbehind window `d`
per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
It is expected that the `series_selector` returns time series of [counter type](https://docs.victoriametrics.com/keyconcepts/#counter).
Unlike Prometheus, it takes into account the last sample before the given lookbehind window `d` when calculating the result.
See [this article](https://medium.com/@romanhavronenko/victoriametrics-promql-compliance-d4318203f51e) for details.
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
This function is supported by PromQL.
See also [increase_pure](#increase_pure), [increase_prometheus](#increase_prometheus) and [delta](#delta).
@@ -529,13 +510,12 @@ See also [increase_pure](#increase_pure), [increase_prometheus](#increase_promet
`increase_prometheus(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the increase
over the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
It is expected that the `series_selector` returns time series of [counter type](https://docs.victoriametrics.com/keyconcepts/#counter).
It doesn't take into account the last sample before the given lookbehind window `d` when calculating the result in the same way as Prometheus does.
See [this article](https://medium.com/@romanhavronenko/victoriametrics-promql-compliance-d4318203f51e) for details.
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
See also [increase_pure](#increase_pure) and [increase](#increase).
#### increase_pure
@@ -544,10 +524,6 @@ See also [increase_pure](#increase_pure) and [increase](#increase).
of the following corner case - it assumes that [counters](https://docs.victoriametrics.com/keyconcepts/#counter) always start from 0,
while [increase](#increase) ignores the first value in a series if it is too big.
This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
See also [increase](#increas) and [increase_prometheus](#increase_prometheus).
#### increases_over_time
`increases_over_time(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the number of [raw sample](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
@@ -564,18 +540,15 @@ on the given lookbehind window `d` per each time series returned from the given
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
#### irate
`irate(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the "instant" per-second increase rate over
the last two [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
on the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
It is expected that the `series_selector` returns time series of [counter type](https://docs.victoriametrics.com/keyconcepts/#counter).
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
This function is supported by PromQL.
See also [rate](#rate) and [rollup_rate](#rollup_rate).
@@ -614,8 +587,6 @@ See also [duration_over_time](#duration_over_time) and [lag](#lag).
over [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples) on the given lookbehind window `d` per each time series returned
from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [mad](#mad), [range_mad](#range_mad) and [outlier_iqr_over_time](#outlier_iqr_over_time).
#### max_over_time
@@ -623,11 +594,9 @@ See also [mad](#mad), [range_mad](#range_mad) and [outlier_iqr_over_time](#outli
`max_over_time(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the maximum value over [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
on the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
This function is supported by PromQL.
See also [tmax_over_time](#tmax_over_time) and [min_over_time](#min_over_time).
See also [tmax_over_time](#tmax_over_time).
#### median_over_time
@@ -635,8 +604,6 @@ See also [tmax_over_time](#tmax_over_time) and [min_over_time](#min_over_time).
on the given lookbehind window `d` per each time series returned
from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [avg_over_time](#avg_over_time).
#### min_over_time
@@ -644,11 +611,9 @@ See also [avg_over_time](#avg_over_time).
`min_over_time(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the minimum value over [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
on the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
This function is supported by PromQL.
See also [tmin_over_time](#tmin_over_time) and [max_over_time](#max_over_time).
See also [tmin_over_time](#tmin_over_time).
#### mode_over_time
@@ -657,8 +622,6 @@ for [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples) on
from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering). It is expected that [raw sample](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
values are discrete.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
#### outlier_iqr_over_time
`outlier_iqr_over_time(series_selector[d])` is a [rollup function](#rollup-functions), which returns the last sample on the given lookbehind window `d`
@@ -669,8 +632,6 @@ if its value is either smaller than the `q25-1.5*iqr` or bigger than `q75+1.5*iq
The `outlier_iqr_over_time()` is useful for detecting anomalies in gauge values based on the previous history of values.
For example, `outlier_iqr_over_time(memory_usage_bytes[1h])` triggers when `memory_usage_bytes` suddenly goes outside the usual value range for the last hour.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [outliers_iqr](#outliers_iqr).
#### predict_linear
@@ -698,8 +659,6 @@ This function is supported by PromQL.
on the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
The `phi` value must be in the range `[0...1]`.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
This function is supported by PromQL.
See also [quantiles_over_time](#quantiles_over_time).
@@ -711,8 +670,6 @@ over [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples) on
from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
The function returns individual series per each `phi*` with `{phiLabel="phi*"}` label. `phi*` values must be in the range `[0...1]`.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [quantile_over_time](#quantile_over_time).
#### range_over_time
@@ -723,12 +680,11 @@ E.g. it calculates `max_over_time(series_selector[d]) - min_over_time(series_sel
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
#### rate
`rate(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the average per-second increase rate
over the given lookbehind window `d` per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
It is expected that the `series_selector` returns time series of [counter type](https://docs.victoriametrics.com/keyconcepts/#counter).
If the lookbehind window is skipped in square brackets, then it is automatically calculated as `max(step, scrape_interval)`, where `step` is the query arg value
passed to [/api/v1/query_range](https://docs.victoriametrics.com/keyconcepts/#range-query) or [/api/v1/query](https://docs.victoriametrics.com/keyconcepts/#instant-query),
@@ -747,8 +703,6 @@ See also [irate](#irate) and [rollup_rate](#rollup_rate).
on the given lookbehind window `d`. The calculations are performed individually per each time series returned
from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
#### resets
@@ -756,11 +710,10 @@ Metric names are stripped from the resulting rollups. Add [keep_metric_names](#k
`resets(series_selector[d])` is a [rollup function](#rollup-functions), which returns the number
of [counter](https://docs.victoriametrics.com/keyconcepts/#counter) resets over the given lookbehind window `d`
per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
It is expected that the `series_selector` returns time series of [counter type](https://docs.victoriametrics.com/keyconcepts/#counter).
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
This function is supported by PromQL.
#### rollup
@@ -772,10 +725,6 @@ These values are calculated individually per each time series returned from the
Optional 2nd argument `"min"`, `"max"` or `"avg"` can be passed to keep only one calculation result and without adding a label.
See also [label_match](#label_match).
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [rollup_rate](#rollup_rate).
#### rollup_candlestick
`rollup_candlestick(series_selector[d])` is a [rollup function](#rollup-functions), which calculates `open`, `high`, `low` and `close` values (aka OHLC)
@@ -787,8 +736,6 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
Optional 2nd argument `"open"`, `"high"` or `"low"` or `"close"` can be passed to keep only one calculation result and without adding a label.
See also [label_match](#label_match).
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
#### rollup_delta
`rollup_delta(series_selector[d])` is a [rollup function](#rollup-functions), which calculates differences between adjacent [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
@@ -815,8 +762,6 @@ See also [label_match](#label_match).
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
See also [rollup](#rollup) and [rollup_rate](#rollup_rate).
#### rollup_increase
`rollup_increase(series_selector[d])` is a [rollup function](#rollup-functions), which calculates increases for adjacent [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
@@ -829,17 +774,12 @@ See also [label_match](#label_match).
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names. See also [rollup_delta](#rollup_delta).
This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
See also [rollup](#rollup) and [rollup_rate](#rollup_rate).
#### rollup_rate
`rollup_rate(series_selector[d])` is a [rollup function](#rollup-functions), which calculates per-second change rates
for adjacent [raw samples](https://docs.victoriametrics.com/keyconcepts/#raw-samples)
on the given lookbehind window `d` and returns `min`, `max` and `avg` values for the calculated per-second change rates
and returns them in time series with `rollup="min"`, `rollup="max"` and `rollup="avg"` additional labels.
The calculations are performed individually per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
See [this article](https://valyala.medium.com/why-irate-from-prometheus-doesnt-capture-spikes-45f9896d7832) in order to understand better
when to use `rollup_rate()`.
@@ -847,12 +787,10 @@ when to use `rollup_rate()`.
Optional 2nd argument `"min"`, `"max"` or `"avg"` can be passed to keep only one calculation result and without adding a label.
See also [label_match](#label_match).
The calculations are performed individually per each time series returned from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#filtering).
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [counters](https://docs.victoriametrics.com/keyconcepts/#counter).
See also [rollup](#rollup) and [rollup_increase](#rollup_increase).
#### rollup_scrape_interval
`rollup_scrape_interval(series_selector[d])` is a [rollup function](#rollup-functions), which calculates the interval in seconds between
@@ -886,8 +824,6 @@ This function is useful for calculating SLI and SLO. Example: `share_gt_over_tim
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [share_le_over_time](#share_le_over_time) and [count_gt_over_time](#count_gt_over_time).
#### share_le_over_time
@@ -902,8 +838,6 @@ the share of time series values for the last 24 hours when memory usage was belo
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [share_gt_over_time](#share_gt_over_time) and [count_le_over_time](#count_le_over_time).
#### share_eq_over_time
@@ -915,8 +849,6 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [count_eq_over_time](#count_eq_over_time).
#### stale_samples_over_time
@@ -934,8 +866,6 @@ on the given lookbehind window `d` per each time series returned from the given
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
This function is supported by PromQL.
See also [stdvar_over_time](#stdvar_over_time).
@@ -947,8 +877,6 @@ on the given lookbehind window `d` per each time series returned from the given
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
This function is supported by PromQL.
See also [stddev_over_time](#stddev_over_time).
@@ -960,8 +888,6 @@ values equal to `eq` on the given lookbehind window `d` per each time series ret
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [sum_over_time](#sum_over_time) and [count_eq_over_time](#count_eq_over_time).
#### sum_gt_over_time
@@ -971,8 +897,6 @@ values bigger than `gt` on the given lookbehind window `d` per each time series
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [sum_over_time](#sum_over_time) and [count_gt_over_time](#count_gt_over_time).
#### sum_le_over_time
@@ -982,8 +906,6 @@ values smaller or equal to `le` on the given lookbehind window `d` per each time
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [sum_over_time](#sum_over_time) and [count_le_over_time](#count_le_over_time).
#### sum_over_time
@@ -993,8 +915,6 @@ on the given lookbehind window `d` per each time series returned from the given
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
This function is supported by PromQL.
#### sum2_over_time
@@ -1004,8 +924,6 @@ values on the given lookbehind window `d` per each time series returned from the
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
#### timestamp
`timestamp(series_selector[d])` is a [rollup function](#rollup-functions), which returns the timestamp in seconds with millisecond precision
@@ -1083,8 +1001,6 @@ from the given [series_selector](https://docs.victoriametrics.com/keyconcepts/#f
Metric names are stripped from the resulting rollups. Add [keep_metric_names](#keep_metric_names) modifier in order to keep metric names.
This function is usually applied to [gauges](https://docs.victoriametrics.com/keyconcepts/#gauge).
See also [zscore](#zscore), [range_trim_zscore](#range_trim_zscore) and [outlier_iqr_over_time](#outlier_iqr_over_time).
@@ -2314,8 +2230,7 @@ Any [rollup function](#rollup-functions) for something other than [series select
Nested rollup functions can be implicit thanks to the [implicit query conversions](#implicit-query-conversions).
For example, `delta(sum(m))` is implicitly converted to `delta(sum(default_rollup(m))[1i:1i])`, so it becomes a subquery,
since it contains [default_rollup](#default_rollup) nested into [delta](#delta).
This behavior can be disabled or logged via `-search.disableImplicitConversion` and `-search.logImplicitConversion` command-line flags
starting from [`v1.101.0` release](https://docs.victoriametrics.com/changelog/).
This behavior can be disabled or logged via cmd-line flags `-search.disableImplicitConversion` and `-search.logImplicitConversion` since v1.101.0.
VictoriaMetrics performs subqueries in the following way:
@@ -2348,6 +2263,5 @@ VictoriaMetrics performs the following implicit conversions for incoming queries
For example, `avg_over_time(rate(http_requests_total[5m])[1h])` is automatically converted to `avg_over_time(rate(http_requests_total[5m])[1h:1i])`.
* If something other than [series selector](https://docs.victoriametrics.com/keyconcepts/#filtering)
is passed to [rollup function](#rollup-functions), then a [subquery](#subqueries) with `1i` lookbehind window and `1i` step is automatically formed.
For example, `rate(sum(up))` is automatically converted to `rate((sum(default_rollup(up)))[1i:1i])`.
This behavior can be disabled or logged via `-search.disableImplicitConversion` and `-search.logImplicitConversion` command-line flags
starting from [`v1.101.0` release](https://docs.victoriametrics.com/changelog/).
For example, `rate(sum(up))` is automatically converted to `rate((sum(default_rollup(up)))[1i:1i])`.
This behavior can be disabled or logged via cmd-line flags `-search.disableImplicitConversion` and `-search.logImplicitConversion` since v1.101.0.

View File

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

View File

@@ -10,8 +10,6 @@ import (
"strings"
"time"
"github.com/VictoriaMetrics/metrics"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/csvimport"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/datadogsketches"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/datadogv1"
@@ -26,6 +24,7 @@ import (
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/prometheusimport"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/promremotewrite"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/remotewrite"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/statsd"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmagent/vmimport"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/auth"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/buildinfo"
@@ -38,13 +37,14 @@ import (
influxserver "github.com/VictoriaMetrics/VictoriaMetrics/lib/ingestserver/influx"
opentsdbserver "github.com/VictoriaMetrics/VictoriaMetrics/lib/ingestserver/opentsdb"
opentsdbhttpserver "github.com/VictoriaMetrics/VictoriaMetrics/lib/ingestserver/opentsdbhttp"
statsdserver "github.com/VictoriaMetrics/VictoriaMetrics/lib/ingestserver/statsd"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/procutil"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/promscrape"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/common"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentelemetry/firehose"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/pushmetrics"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/stringsutil"
"github.com/VictoriaMetrics/metrics"
)
var (
@@ -63,6 +63,12 @@ var (
"See also -graphiteListenAddr.useProxyProtocol")
graphiteUseProxyProtocol = flag.Bool("graphiteListenAddr.useProxyProtocol", false, "Whether to use proxy protocol for connections accepted at -graphiteListenAddr . "+
"See https://www.haproxy.org/download/1.8/doc/proxy-protocol.txt")
statsdListenAddr = flag.String("statsdListenAddr", "", "TCP and UDP address to listen for Statsd plaintext data. Usually :8125 must be set. Doesn't work if empty. "+
"See also -statsdListenAddr.useProxyProtocol")
statsdUseProxyProtocol = flag.Bool("statsdListenAddr.useProxyProtocol", false, "Whether to use proxy protocol for connections accepted at -statsdListenAddr . "+
"See https://www.haproxy.org/download/1.8/doc/proxy-protocol.txt")
statsdDisableAggregationEnforcemenet = flag.Bool(`statsd.disableAggregationEnforcement`, false, "Whether to disable streaming aggregation requirement check. "+
"It's recommended to run statsdServer with pre-configured streaming aggregation to decrease load at database.")
opentsdbListenAddr = flag.String("opentsdbListenAddr", "", "TCP and UDP address to listen for OpenTSDB metrics. "+
"Telnet put messages and HTTP /api/put messages are simultaneously served on TCP port. "+
"Usually :4242 must be set. Doesn't work if empty. See also -opentsdbListenAddr.useProxyProtocol")
@@ -72,8 +78,8 @@ var (
"See also -opentsdbHTTPListenAddr.useProxyProtocol")
opentsdbHTTPUseProxyProtocol = flag.Bool("opentsdbHTTPListenAddr.useProxyProtocol", false, "Whether to use proxy protocol for connections accepted "+
"at -opentsdbHTTPListenAddr . See https://www.haproxy.org/download/1.8/doc/proxy-protocol.txt")
configAuthKey = flagutil.NewPassword("configAuthKey", "Authorization key for accessing /config page. It must be passed via authKey query arg. It overrides -httpAuth.*")
reloadAuthKey = flagutil.NewPassword("reloadAuthKey", "Auth key for /-/reload http endpoint. It must be passed via authKey query arg. It overrides -httpAuth.*")
configAuthKey = flagutil.NewPassword("configAuthKey", "Authorization key for accessing /config page. It must be passed via authKey query arg. It overrides httpAuth.* settings.")
reloadAuthKey = flagutil.NewPassword("reloadAuthKey", "Auth key for /-/reload http endpoint. It must be passed via authKey query arg. It overrides httpAuth.* settings.")
dryRun = flag.Bool("dryRun", false, "Whether to check config files without running vmagent. The following files are checked: "+
"-promscrape.config, -remoteWrite.relabelConfig, -remoteWrite.urlRelabelConfig, -remoteWrite.streamAggr.config . "+
"Unknown config entries aren't allowed in -promscrape.config by default. This can be changed by passing -promscrape.config.strictParse=false command-line flag")
@@ -82,6 +88,7 @@ var (
var (
influxServer *influxserver.Server
graphiteServer *graphiteserver.Server
statsdServer *statsdserver.Server
opentsdbServer *opentsdbserver.Server
opentsdbhttpServer *opentsdbhttpserver.Server
)
@@ -116,7 +123,7 @@ func main() {
logger.Fatalf("error when checking relabel configs: %s", err)
}
if err := remotewrite.CheckStreamAggrConfigs(); err != nil {
logger.Fatalf("error when checking -streamAggr.config and -remoteWrite.streamAggr.config: %s", err)
logger.Fatalf("error when checking -remoteWrite.streamAggr.config: %s", err)
}
logger.Infof("all the configs are ok; exiting with 0 status code")
return
@@ -139,6 +146,12 @@ func main() {
if len(*graphiteListenAddr) > 0 {
graphiteServer = graphiteserver.MustStart(*graphiteListenAddr, *graphiteUseProxyProtocol, graphite.InsertHandler)
}
if len(*statsdListenAddr) > 0 {
if !remotewrite.HasAnyStreamAggrConfigured() && !*statsdDisableAggregationEnforcemenet {
logger.Fatalf("streaming aggregation must be configured with enabled statsd server. It's recommended to aggregate metrics received at statsd listener. This check could be disabled with flag -statsd.disableAggregationEnforcement")
}
statsdServer = statsdserver.MustStart(*statsdListenAddr, *statsdUseProxyProtocol, statsd.InsertHandler)
}
if len(*opentsdbListenAddr) > 0 {
httpInsertHandler := getOpenTSDBHTTPInsertHandler()
opentsdbServer = opentsdbserver.MustStart(*opentsdbListenAddr, *opentsdbUseProxyProtocol, opentsdb.InsertHandler, httpInsertHandler)
@@ -174,6 +187,9 @@ func main() {
if len(*graphiteListenAddr) > 0 {
graphiteServer.MustStop()
}
if len(*statsdListenAddr) > 0 {
statsdServer.MustStop()
}
if len(*opentsdbListenAddr) > 0 {
opentsdbServer.MustStop()
}
@@ -318,10 +334,6 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
influxQueryRequests.Inc()
influxutils.WriteDatabaseNames(w)
return true
case "/influx/health":
influxHealthRequests.Inc()
influxutils.WriteHealthCheckResponse(w)
return true
case "/opentelemetry/api/v1/push", "/opentelemetry/v1/metrics":
opentelemetryPushRequests.Inc()
if err := opentelemetry.InsertHandler(nil, r); err != nil {
@@ -440,7 +452,7 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
}
return true
case "/prometheus/config", "/config":
if !httpserver.CheckAuthFlag(w, r, configAuthKey) {
if !httpserver.CheckAuthFlag(w, r, configAuthKey.Get(), "configAuthKey") {
return true
}
promscrapeConfigRequests.Inc()
@@ -449,17 +461,17 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
return true
case "/prometheus/api/v1/status/config", "/api/v1/status/config":
// See https://prometheus.io/docs/prometheus/latest/querying/api/#config
if !httpserver.CheckAuthFlag(w, r, configAuthKey) {
if !httpserver.CheckAuthFlag(w, r, configAuthKey.Get(), "configAuthKey") {
return true
}
promscrapeStatusConfigRequests.Inc()
w.Header().Set("Content-Type", "application/json")
var bb bytesutil.ByteBuffer
promscrape.WriteConfigData(&bb)
fmt.Fprintf(w, `{"status":"success","data":{"yaml":%s}}`, stringsutil.JSONString(string(bb.B)))
fmt.Fprintf(w, `{"status":"success","data":{"yaml":%q}}`, bb.B)
return true
case "/prometheus/-/reload", "/-/reload":
if !httpserver.CheckAuthFlag(w, r, reloadAuthKey) {
if !httpserver.CheckAuthFlag(w, r, reloadAuthKey.Get(), "reloadAuthKey") {
return true
}
promscrapeConfigReloadRequests.Inc()
@@ -568,10 +580,6 @@ func processMultitenantRequest(w http.ResponseWriter, r *http.Request, path stri
influxQueryRequests.Inc()
influxutils.WriteDatabaseNames(w)
return true
case "influx/health":
influxHealthRequests.Inc()
influxutils.WriteHealthCheckResponse(w)
return true
case "opentelemetry/api/v1/push", "opentelemetry/v1/metrics":
opentelemetryPushRequests.Inc()
if err := opentelemetry.InsertHandler(at, r); err != nil {
@@ -682,8 +690,7 @@ var (
influxWriteRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/influx/write", protocol="influx"}`)
influxWriteErrors = metrics.NewCounter(`vmagent_http_request_errors_total{path="/influx/write", protocol="influx"}`)
influxQueryRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/influx/query", protocol="influx"}`)
influxHealthRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/influx/health", protocol="influx"}`)
influxQueryRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/influx/query", protocol="influx"}`)
datadogv1WriteRequests = metrics.NewCounter(`vmagent_http_requests_total{path="/datadog/api/v1/series", protocol="datadog"}`)
datadogv1WriteErrors = metrics.NewCounter(`vmagent_http_request_errors_total{path="/datadog/api/v1/series", protocol="datadog"}`)

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

@@ -6,7 +6,6 @@ import (
"net/http"
"net/url"
"path/filepath"
"slices"
"strconv"
"sync"
"sync/atomic"
@@ -89,15 +88,15 @@ var (
"By default there are no limits on samples ingestion rate. See also -remoteWrite.rateLimit")
disableOnDiskQueue = flagutil.NewArrayBool("remoteWrite.disableOnDiskQueue", "Whether to disable storing pending data to -remoteWrite.tmpDataPath "+
"when the remote storage system at the corresponding -remoteWrite.url cannot keep up with the data ingestion rate. "+
"See https://docs.victoriametrics.com/vmagent#disabling-on-disk-persistence . See also -remoteWrite.dropSamplesOnOverload")
dropSamplesOnOverload = flag.Bool("remoteWrite.dropSamplesOnOverload", false, "Whether to drop samples when -remoteWrite.disableOnDiskQueue is set and if the samples "+
"cannot be pushed into the configured -remoteWrite.url systems in a timely manner. See https://docs.victoriametrics.com/vmagent#disabling-on-disk-persistence")
"when the configured remote storage systems cannot keep up with the data ingestion rate. See https://docs.victoriametrics.com/vmagent#disabling-on-disk-persistence ."+
"See also -remoteWrite.dropSamplesOnOverload")
dropSamplesOnOverload = flagutil.NewArrayBool("remoteWrite.dropSamplesOnOverload", "Whether to drop samples when -remoteWrite.disableOnDiskQueue is set and if the samples "+
"cannot be pushed into the configured remote storage systems in a timely manner. See https://docs.victoriametrics.com/vmagent#disabling-on-disk-persistence")
)
var (
// rwctxsGlobal contains statically populated entries when -remoteWrite.url is specified.
rwctxsGlobal []*remoteWriteCtx
// rwctxs contains statically populated entries when -remoteWrite.url is specified.
rwctxs []*remoteWriteCtx
// Data without tenant id is written to defaultAuthToken if -enableMultitenantHandlers is specified.
defaultAuthToken = &auth.Token{}
@@ -110,11 +109,8 @@ var (
StatusCode: http.StatusTooManyRequests,
}
// disableOnDiskQueueAny is set to true if at least a single -remoteWrite.url is configured with -remoteWrite.disableOnDiskQueue
disableOnDiskQueueAny bool
// dropSamplesOnFailureGlobal is set to true if -remoteWrite.dropSamplesOnOverload is set or if multiple -remoteWrite.disableOnDiskQueue options are set.
dropSamplesOnFailureGlobal bool
// disableOnDiskQueueAll is set to true if all remoteWrite.urls were configured to disable persistent queue via disableOnDiskQueue
disableOnDiskQueueAll bool
)
// MultitenancyEnabled returns true if -enableMultitenantHandlers is specified.
@@ -207,18 +203,28 @@ func Init() {
relabelConfigSuccess.Set(1)
relabelConfigTimestamp.Set(fasttime.UnixTimestamp())
initStreamAggrConfigGlobal()
sasFile, sasOpts := getStreamAggrOpts(-1)
if sasFile != "" {
sas, err := newStreamAggrConfig(-1, pushToRemoteStoragesDropFailed)
if err != nil {
logger.Fatalf("cannot initialize stream aggregators from -streamAggr.config=%q: %s", sasFile, err)
}
sasGlobal.Store(sas)
} else if sasOpts.DedupInterval > 0 {
deduplicatorGlobal = streamaggr.NewDeduplicator(pushToRemoteStoragesDropFailed, sasOpts.DedupInterval, sasOpts.DropInputLabels)
}
rwctxsGlobal = newRemoteWriteCtxs(nil, *remoteWriteURLs)
if len(*remoteWriteURLs) > 0 {
rwctxs = newRemoteWriteCtxs(nil, *remoteWriteURLs)
}
disableOnDiskQueues := []bool(*disableOnDiskQueue)
disableOnDiskQueueAny = slices.Contains(disableOnDiskQueues, true)
// Samples must be dropped if multiple -remoteWrite.disableOnDiskQueue options are configured and at least a single is set to true.
// In this case it is impossible to prevent from sending many duplicates of samples passed to TryPush() to all the configured -remoteWrite.url
// if these samples couldn't be sent to the -remoteWrite.url with the disabled persistent queue. So it is better sending samples
// to the remaining -remoteWrite.url and dropping them on the blocked queue.
dropSamplesOnFailureGlobal = *dropSamplesOnOverload || disableOnDiskQueueAny && len(disableOnDiskQueues) > 1
disableOnDiskQueueAll = true
for _, v := range *disableOnDiskQueue {
if !v {
disableOnDiskQueueAll = false
break
}
}
dropDanglingQueues()
@@ -228,9 +234,9 @@ func Init() {
defer configReloaderWG.Done()
for {
select {
case <-sighupCh:
case <-configReloaderStopCh:
return
case <-sighupCh:
}
reloadRelabelConfigs()
reloadStreamAggrConfigs()
@@ -249,8 +255,8 @@ func dropDanglingQueues() {
// In case if there were many persistent queues with identical *remoteWriteURLs
// the queue with the last index will be dropped.
// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/6140
existingQueues := make(map[string]struct{}, len(rwctxsGlobal))
for _, rwctx := range rwctxsGlobal {
existingQueues := make(map[string]struct{}, len(rwctxs))
for _, rwctx := range rwctxs {
existingQueues[rwctx.fq.Dirname()] = struct{}{}
}
@@ -267,7 +273,7 @@ func dropDanglingQueues() {
}
}
if removed > 0 {
logger.Infof("removed %d dangling queues from %q, active queues: %d", removed, *tmpDataPath, len(rwctxsGlobal))
logger.Infof("removed %d dangling queues from %q, active queues: %d", removed, *tmpDataPath, len(rwctxs))
}
}
@@ -376,10 +382,10 @@ func Stop() {
deduplicatorGlobal = nil
}
for _, rwctx := range rwctxsGlobal {
for _, rwctx := range rwctxs {
rwctx.MustStop()
}
rwctxsGlobal = nil
rwctxs = nil
if sl := hourlySeriesLimiter; sl != nil {
sl.MustStop()
@@ -391,8 +397,6 @@ func Stop() {
// PushDropSamplesOnFailure pushes wr to the configured remote storage systems set via -remoteWrite.url
//
// PushDropSamplesOnFailure drops wr samples if they cannot be sent to -remoteWrite.url by any reason.
//
// PushDropSamplesOnFailure can modify wr contents.
func PushDropSamplesOnFailure(at *auth.Token, wr *prompbmarshal.WriteRequest) {
_ = tryPush(at, wr, true)
@@ -405,7 +409,7 @@ func PushDropSamplesOnFailure(at *auth.Token, wr *prompbmarshal.WriteRequest) {
//
// The caller must return ErrQueueFullHTTPRetry to the client, which sends wr, if TryPush returns false.
func TryPush(at *auth.Token, wr *prompbmarshal.WriteRequest) bool {
return tryPush(at, wr, dropSamplesOnFailureGlobal)
return tryPush(at, wr, false)
}
func tryPush(at *auth.Token, wr *prompbmarshal.WriteRequest, forceDropSamplesOnFailure bool) bool {
@@ -422,30 +426,33 @@ func tryPush(at *auth.Token, wr *prompbmarshal.WriteRequest, forceDropSamplesOnF
tenantRctx = getRelabelCtx()
defer putRelabelCtx(tenantRctx)
}
rowsCount := getRowsCount(tss)
// Quick check whether writes to configured remote storage systems are blocked.
// This allows saving CPU time spent on relabeling and block compression
// if some of remote storage systems cannot keep up with the data ingestion rate.
rwctxs, ok := getEligibleRemoteWriteCtxs(tss, forceDropSamplesOnFailure)
if !ok {
// At least a single remote write queue is blocked and dropSamplesOnFailure isn't set.
// Return false to the caller, so it could re-send samples again.
return false
}
if len(rwctxs) == 0 {
// All the remote write queues are skipped because they are blocked and dropSamplesOnFailure is set to true.
// Return true to the caller, so it doesn't re-send the samples again.
return true
// this shortcut is only applicable if all remote writes have disableOnDiskQueue = true
if disableOnDiskQueueAll {
for _, rwctx := range rwctxs {
if rwctx.fq.IsWriteBlocked() {
rwctx.pushFailures.Inc()
if forceDropSamplesOnFailure || rwctx.dropSamplesOnOverload {
// Just drop samples
rwctx.rowsDroppedOnPushFailure.Add(rowsCount)
continue
}
return false
}
}
}
var rctx *relabelCtx
rcs := allRelabelConfigs.Load()
pcsGlobal := rcs.global
if pcsGlobal.Len() > 0 || *usePromCompatibleNaming {
if pcsGlobal.Len() > 0 {
rctx = getRelabelCtx()
defer putRelabelCtx(rctx)
}
rowsCount := getRowsCount(tss)
globalRowsPushedBeforeRelabel.Add(rowsCount)
maxSamplesPerBlock := *maxRowsPerBlock
// Allow up to 10x of labels per each block on average.
@@ -498,46 +505,20 @@ func tryPush(at *auth.Token, wr *prompbmarshal.WriteRequest, forceDropSamplesOnF
deduplicatorGlobal.Push(tssBlock)
tssBlock = tssBlock[:0]
}
if !tryPushBlockToRemoteStorages(rwctxs, tssBlock, forceDropSamplesOnFailure) {
if !tryPushBlockToRemoteStorages(tssBlock, forceDropSamplesOnFailure) {
return false
}
}
return true
}
func getEligibleRemoteWriteCtxs(tss []prompbmarshal.TimeSeries, forceDropSamplesOnFailure bool) ([]*remoteWriteCtx, bool) {
if !disableOnDiskQueueAny {
return rwctxsGlobal, true
}
// This code is applicable if at least a single remote storage has -disableOnDiskQueue
rwctxs := make([]*remoteWriteCtx, 0, len(rwctxsGlobal))
for _, rwctx := range rwctxsGlobal {
if !rwctx.fq.IsWriteBlocked() {
rwctxs = append(rwctxs, rwctx)
} else {
rwctx.pushFailures.Inc()
if !forceDropSamplesOnFailure {
return nil, false
}
rowsCount := getRowsCount(tss)
rwctx.rowsDroppedOnPushFailure.Add(rowsCount)
}
}
return rwctxs, true
}
func pushToRemoteStoragesTrackDropped(tss []prompbmarshal.TimeSeries) {
rwctxs, _ := getEligibleRemoteWriteCtxs(tss, true)
if len(rwctxs) == 0 {
func pushToRemoteStoragesDropFailed(tss []prompbmarshal.TimeSeries) {
if tryPushBlockToRemoteStorages(tss, true) {
return
}
if !tryPushBlockToRemoteStorages(rwctxs, tss, true) {
logger.Panicf("BUG: tryPushBlockToRemoteStorages() must return true when forceDropSamplesOnFailure=true")
}
}
func tryPushBlockToRemoteStorages(rwctxs []*remoteWriteCtx, tssBlock []prompbmarshal.TimeSeries, forceDropSamplesOnFailure bool) bool {
func tryPushBlockToRemoteStorages(tssBlock []prompbmarshal.TimeSeries, forceDropSamplesOnFailure bool) bool {
if len(tssBlock) == 0 {
// Nothing to push
return true
@@ -556,7 +537,7 @@ func tryPushBlockToRemoteStorages(rwctxs []*remoteWriteCtx, tssBlock []prompbmar
if replicas <= 0 {
replicas = 1
}
return tryShardingBlockAmongRemoteStorages(rwctxs, tssBlock, replicas, forceDropSamplesOnFailure)
return tryShardingBlockAmongRemoteStorages(tssBlock, replicas, forceDropSamplesOnFailure)
}
// Replicate tssBlock samples among rwctxs.
@@ -577,7 +558,7 @@ func tryPushBlockToRemoteStorages(rwctxs []*remoteWriteCtx, tssBlock []prompbmar
return !anyPushFailed.Load()
}
func tryShardingBlockAmongRemoteStorages(rwctxs []*remoteWriteCtx, tssBlock []prompbmarshal.TimeSeries, replicas int, forceDropSamplesOnFailure bool) bool {
func tryShardingBlockAmongRemoteStorages(tssBlock []prompbmarshal.TimeSeries, replicas int, forceDropSamplesOnFailure bool) bool {
x := getTSSShards(len(rwctxs))
defer putTSSShards(x)
@@ -764,8 +745,10 @@ type remoteWriteCtx struct {
sas atomic.Pointer[streamaggr.Aggregators]
deduplicator *streamaggr.Deduplicator
streamAggrKeepInput bool
streamAggrDropInput bool
streamAggrKeepInput bool
streamAggrDropInput bool
disableOnDiskQueue bool
dropSamplesOnOverload bool
pss []*pendingSeries
pssNextIdx atomic.Uint64
@@ -790,7 +773,6 @@ func newRemoteWriteCtx(argIdx int, remoteWriteURL *url.URL, maxInmemoryBlocks in
logger.Warnf("rounding the -remoteWrite.maxDiskUsagePerURL=%d to the minimum supported value: %d", maxPendingBytes, persistentqueue.DefaultChunkFileSize)
maxPendingBytes = persistentqueue.DefaultChunkFileSize
}
isPQDisabled := disableOnDiskQueue.GetOptionalArg(argIdx)
fq := persistentqueue.MustOpenFastQueue(queuePath, sanitizedURL, maxInmemoryBlocks, maxPendingBytes, isPQDisabled)
_ = metrics.GetOrCreateGauge(fmt.Sprintf(`vmagent_remotewrite_pending_data_bytes{path=%q, url=%q}`, queuePath, sanitizedURL), func() float64 {
@@ -835,20 +817,38 @@ func newRemoteWriteCtx(argIdx int, remoteWriteURL *url.URL, maxInmemoryBlocks in
c: c,
pss: pss,
rowsPushedAfterRelabel: metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_rows_pushed_after_relabel_total{path=%q,url=%q}`, queuePath, sanitizedURL)),
rowsDroppedByRelabel: metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_relabel_metrics_dropped_total{path=%q,url=%q}`, queuePath, sanitizedURL)),
dropSamplesOnOverload: dropSamplesOnOverload.GetOptionalArg(argIdx),
disableOnDiskQueue: isPQDisabled,
pushFailures: metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_push_failures_total{path=%q,url=%q}`, queuePath, sanitizedURL)),
rowsDroppedOnPushFailure: metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_samples_dropped_total{path=%q,url=%q}`, queuePath, sanitizedURL)),
rowsPushedAfterRelabel: metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_rows_pushed_after_relabel_total{path=%q, url=%q}`, queuePath, sanitizedURL)),
rowsDroppedByRelabel: metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_relabel_metrics_dropped_total{path=%q, url=%q}`, queuePath, sanitizedURL)),
pushFailures: metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_push_failures_total{path=%q, url=%q}`, queuePath, sanitizedURL)),
rowsDroppedOnPushFailure: metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_samples_dropped_total{path=%q, url=%q}`, queuePath, sanitizedURL)),
}
// Initialize sas
sasFile, sasOpts := getStreamAggrOpts(argIdx)
if sasFile != "" {
sas, err := newStreamAggrConfig(argIdx, rwctx.pushInternalTrackDropped)
if err != nil {
logger.Fatalf("cannot initialize stream aggregators from -remoteWrite.streamAggr.config=%q: %s", sasFile, err)
}
rwctx.sas.Store(sas)
rwctx.streamAggrKeepInput = streamAggrKeepInput.GetOptionalArg(argIdx)
rwctx.streamAggrDropInput = streamAggrDropInput.GetOptionalArg(argIdx)
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, sasFile)).Set(1)
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_success_timestamp_seconds{path=%q}`, sasFile)).Set(fasttime.UnixTimestamp())
} else if sasOpts.DedupInterval > 0 {
rwctx.deduplicator = streamaggr.NewDeduplicator(rwctx.pushInternalTrackDropped, sasOpts.DedupInterval, sasOpts.DropInputLabels)
}
rwctx.initStreamAggrConfig()
return rwctx
}
func (rwctx *remoteWriteCtx) MustStop() {
// sas and deduplicator must be stopped before rwctx is closed
// because they can write pending series to rwctx.pss if there are any
// because sas can write pending series to rwctx.pss if there are any
sas := rwctx.sas.Swap(nil)
sas.MustStop()
@@ -875,20 +875,12 @@ func (rwctx *remoteWriteCtx) MustStop() {
// TryPush sends tss series to the configured remote write endpoint
//
// TryPush doesn't modify tss, so tss can be passed concurrently to TryPush across distinct rwctx instances.
// TryPush can be called concurrently for multiple remoteWriteCtx,
// so it shouldn't modify tss entries.
func (rwctx *remoteWriteCtx) TryPush(tss []prompbmarshal.TimeSeries, forceDropSamplesOnFailure bool) bool {
// Apply relabeling
var rctx *relabelCtx
var v *[]prompbmarshal.TimeSeries
defer func() {
if rctx == nil {
return
}
*v = prompbmarshal.ResetTimeSeries(tss)
tssPool.Put(v)
putRelabelCtx(rctx)
}()
// Apply relabeling
rcs := allRelabelConfigs.Load()
pcs := rcs.perURL[rwctx.idx]
if pcs.Len() > 0 {
@@ -924,22 +916,28 @@ func (rwctx *remoteWriteCtx) TryPush(tss []prompbmarshal.TimeSeries, forceDropSa
matchIdxsPool.Put(matchIdxs)
} else if rwctx.deduplicator != nil {
rwctx.deduplicator.Push(tss)
return true
tss = tss[:0]
}
// Try pushing tss to remote storage
if rwctx.tryPushInternal(tss) {
return true
// Try pushing the data to remote storage
ok := rwctx.tryPushInternal(tss)
// Return back relabeling contexts to the pool
if rctx != nil {
*v = prompbmarshal.ResetTimeSeries(tss)
tssPool.Put(v)
putRelabelCtx(rctx)
}
// Couldn't push tss to remote storage
rwctx.pushFailures.Inc()
if forceDropSamplesOnFailure {
rowsCount := getRowsCount(tss)
rwctx.rowsDroppedOnPushFailure.Add(rowsCount)
return true
if !ok {
rwctx.pushFailures.Inc()
if forceDropSamplesOnFailure || rwctx.dropSamplesOnOverload {
rwctx.rowsDroppedOnPushFailure.Add(len(tss))
return true
}
}
return false
return ok
}
var matchIdxsPool bytesutil.ByteBufferPool
@@ -963,26 +961,19 @@ func (rwctx *remoteWriteCtx) pushInternalTrackDropped(tss []prompbmarshal.TimeSe
if rwctx.tryPushInternal(tss) {
return
}
if !rwctx.fq.IsPersistentQueueDisabled() {
if !rwctx.disableOnDiskQueue {
logger.Panicf("BUG: tryPushInternal must return true if -remoteWrite.disableOnDiskQueue isn't set")
}
rwctx.pushFailures.Inc()
rowsCount := getRowsCount(tss)
rwctx.rowsDroppedOnPushFailure.Add(rowsCount)
if dropSamplesOnOverload.GetOptionalArg(rwctx.idx) {
rowsCount := getRowsCount(tss)
rwctx.rowsDroppedOnPushFailure.Add(rowsCount)
}
}
func (rwctx *remoteWriteCtx) tryPushInternal(tss []prompbmarshal.TimeSeries) bool {
var rctx *relabelCtx
var v *[]prompbmarshal.TimeSeries
defer func() {
if rctx == nil {
return
}
*v = prompbmarshal.ResetTimeSeries(tss)
tssPool.Put(v)
putRelabelCtx(rctx)
}()
if len(labelsGlobal) > 0 {
// Make a copy of tss before adding extra labels in order to prevent
// from affecting time series for other remoteWrite.url configs.
@@ -995,11 +986,19 @@ func (rwctx *remoteWriteCtx) tryPushInternal(tss []prompbmarshal.TimeSeries) boo
pss := rwctx.pss
idx := rwctx.pssNextIdx.Add(1) % uint64(len(pss))
return pss[idx].TryPush(tss)
ok := pss[idx].TryPush(tss)
if rctx != nil {
*v = prompbmarshal.ResetTimeSeries(tss)
tssPool.Put(v)
putRelabelCtx(rctx)
}
return ok
}
var tssPool = &sync.Pool{
New: func() any {
New: func() interface{} {
a := []prompbmarshal.TimeSeries{}
return &a
},

View File

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

View File

@@ -36,207 +36,125 @@ var (
"before stream de-duplication and aggregation . See https://docs.victoriametrics.com/stream-aggregation/#dropping-unneeded-labels")
// Per URL config
streamAggrConfig = flagutil.NewArrayString("remoteWrite.streamAggr.config", "Optional path to file with stream aggregation config for the corresponding -remoteWrite.url. "+
streamAggrConfig = flagutil.NewArrayString("remoteWrite.streamAggr.config", "Optional path to file with stream aggregation config. "+
"See https://docs.victoriametrics.com/stream-aggregation/ . "+
"See also -remoteWrite.streamAggr.keepInput, -remoteWrite.streamAggr.dropInput and -remoteWrite.streamAggr.dedupInterval")
streamAggrDropInput = flagutil.NewArrayBool("remoteWrite.streamAggr.dropInput", "Whether to drop all the input samples after the aggregation "+
"with -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. By default, only aggregates samples are dropped, while the remaining samples "+
"with -remoteWrite.streamAggr.config. By default, only aggregates samples are dropped, while the remaining samples "+
"are written to the corresponding -remoteWrite.url . See also -remoteWrite.streamAggr.keepInput and https://docs.victoriametrics.com/stream-aggregation/")
streamAggrKeepInput = flagutil.NewArrayBool("remoteWrite.streamAggr.keepInput", "Whether to keep all the input samples after the aggregation "+
"with -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. By default, only aggregates samples are dropped, while the remaining samples "+
"with -remoteWrite.streamAggr.config. By default, only aggregates samples are dropped, while the remaining samples "+
"are written to the corresponding -remoteWrite.url . See also -remoteWrite.streamAggr.dropInput and https://docs.victoriametrics.com/stream-aggregation/")
streamAggrDedupInterval = flagutil.NewArrayDuration("remoteWrite.streamAggr.dedupInterval", 0, "Input samples are de-duplicated with this interval before optional aggregation "+
"with -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. See also -dedup.minScrapeInterval and https://docs.victoriametrics.com/stream-aggregation/#deduplication")
"with -remoteWrite.streamAggr.config . See also -dedup.minScrapeInterval and https://docs.victoriametrics.com/stream-aggregation/#deduplication")
streamAggrIgnoreOldSamples = flagutil.NewArrayBool("remoteWrite.streamAggr.ignoreOldSamples", "Whether to ignore input samples with old timestamps outside the current "+
"aggregation interval for the corresponding -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. "+
"aggregation interval for the corresponding -remoteWrite.streamAggr.config . "+
"See https://docs.victoriametrics.com/stream-aggregation/#ignoring-old-samples")
streamAggrIgnoreFirstIntervals = flagutil.NewArrayInt("remoteWrite.streamAggr.ignoreFirstIntervals", 0, "Number of aggregation intervals to skip after the start "+
"for the corresponding -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. Increase this value if "+
"you observe incorrect aggregation results after vmagent restarts. It could be caused by receiving bufferred delayed data from clients pushing data into the vmagent. "+
streamAggrIgnoreFirstIntervals = flag.Int("remoteWrite.streamAggr.ignoreFirstIntervals", 0, "Number of aggregation intervals to skip after the start. Increase this value if "+
"you observe incorrect aggregation results after vmagent restarts. It could be caused by receiving unordered delayed data from clients pushing data into the vmagent. "+
"See https://docs.victoriametrics.com/stream-aggregation/#ignore-aggregation-intervals-on-start")
streamAggrDropInputLabels = flagutil.NewArrayString("remoteWrite.streamAggr.dropInputLabels", "An optional list of labels to drop from samples "+
"before stream de-duplication and aggregation with -remoteWrite.streamAggr.config and -remoteWrite.streamAggr.dedupInterval at the corresponding -remoteWrite.url. "+
"See https://docs.victoriametrics.com/stream-aggregation/#dropping-unneeded-labels")
"before stream de-duplication and aggregation . See https://docs.victoriametrics.com/stream-aggregation/#dropping-unneeded-labels")
)
// CheckStreamAggrConfigs checks -remoteWrite.streamAggr.config and -streamAggr.config.
func CheckStreamAggrConfigs() error {
// Check global config
sas, err := newStreamAggrConfigGlobal()
if err != nil {
return err
}
sas.MustStop()
if len(*streamAggrConfig) > len(*remoteWriteURLs) {
return fmt.Errorf("too many -remoteWrite.streamAggr.config args: %d; it mustn't exceed the number of -remoteWrite.url args: %d", len(*streamAggrConfig), len(*remoteWriteURLs))
}
pushNoop := func(_ []prompbmarshal.TimeSeries) {}
if _, err := newStreamAggrConfig(-1, pushNoop); err != nil {
return fmt.Errorf("could not load -streamAggr.config stream aggregation config: %w", err)
}
if len(*streamAggrConfig) > len(*remoteWriteURLs) {
return fmt.Errorf("too many -remoteWrite.streamAggr.config args: %d; it mustn't exceed the number of -remoteWrite.url args: %d",
len(*streamAggrConfig), len(*remoteWriteURLs))
}
for idx := range *streamAggrConfig {
sas, err := newStreamAggrConfigPerURL(idx, pushNoop)
if err != nil {
if _, err := newStreamAggrConfig(idx, pushNoop); err != nil {
return err
}
sas.MustStop()
}
return nil
}
// HasAnyStreamAggrConfigured checks if any streaming aggregation config provided
func HasAnyStreamAggrConfigured() bool {
return len(*streamAggrConfig) > 0 || *streamAggrGlobalConfig != ""
}
func reloadStreamAggrConfigs() {
reloadStreamAggrConfigGlobal()
for _, rwctx := range rwctxsGlobal {
rwctx.reloadStreamAggrConfig()
reloadStreamAggrConfig(-1, pushToRemoteStoragesDropFailed)
for idx, rwctx := range rwctxs {
reloadStreamAggrConfig(idx, rwctx.pushInternalTrackDropped)
}
}
func reloadStreamAggrConfigGlobal() {
path := *streamAggrGlobalConfig
if path == "" {
return
}
logger.Infof("reloading stream aggregation configs pointed by -streamAggr.config=%q", path)
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reloads_total{path=%q}`, path)).Inc()
sasNew, err := newStreamAggrConfigGlobal()
if err != nil {
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reloads_errors_total{path=%q}`, path)).Inc()
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, path)).Set(0)
logger.Errorf("cannot reload -streamAggr.config=%q; continue using the previously loaded config; error: %s", path, err)
return
}
sas := sasGlobal.Load()
if !sasNew.Equal(sas) {
sasOld := sasGlobal.Swap(sasNew)
sasOld.MustStop()
logger.Infof("successfully reloaded -streamAggr.config=%q", path)
} else {
sasNew.MustStop()
logger.Infof("-streamAggr.config=%q wasn't changed since the last reload", path)
}
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, path)).Set(1)
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_success_timestamp_seconds{path=%q}`, path)).Set(fasttime.UnixTimestamp())
}
func initStreamAggrConfigGlobal() {
sas, err := newStreamAggrConfigGlobal()
if err != nil {
logger.Fatalf("cannot initialize gloabl stream aggregators: %s", err)
}
if sas != nil {
filePath := sas.FilePath()
sasGlobal.Store(sas)
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, filePath)).Set(1)
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_success_timestamp_seconds{path=%q}`, filePath)).Set(fasttime.UnixTimestamp())
} else {
dedupInterval := streamAggrGlobalDedupInterval.Duration()
if dedupInterval > 0 {
deduplicatorGlobal = streamaggr.NewDeduplicator(pushToRemoteStoragesTrackDropped, dedupInterval, *streamAggrGlobalDropInputLabels, "dedup-global")
}
}
}
func (rwctx *remoteWriteCtx) initStreamAggrConfig() {
idx := rwctx.idx
sas, err := rwctx.newStreamAggrConfig()
if err != nil {
logger.Fatalf("cannot initialize stream aggregators: %s", err)
}
if sas != nil {
filePath := sas.FilePath()
rwctx.sas.Store(sas)
rwctx.streamAggrKeepInput = streamAggrKeepInput.GetOptionalArg(idx)
rwctx.streamAggrDropInput = streamAggrDropInput.GetOptionalArg(idx)
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, filePath)).Set(1)
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_success_timestamp_seconds{path=%q}`, filePath)).Set(fasttime.UnixTimestamp())
} else {
dedupInterval := streamAggrDedupInterval.GetOptionalArg(idx)
if dedupInterval > 0 {
alias := fmt.Sprintf("dedup-%d", idx+1)
rwctx.deduplicator = streamaggr.NewDeduplicator(rwctx.pushInternalTrackDropped, dedupInterval, *streamAggrDropInputLabels, alias)
}
}
}
func (rwctx *remoteWriteCtx) reloadStreamAggrConfig() {
path := streamAggrConfig.GetOptionalArg(rwctx.idx)
if path == "" {
return
}
func reloadStreamAggrConfig(idx int, pushFunc streamaggr.PushFunc) {
path, opts := getStreamAggrOpts(idx)
logger.Infof("reloading stream aggregation configs pointed by -remoteWrite.streamAggr.config=%q", path)
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reloads_total{path=%q}`, path)).Inc()
sasNew, err := rwctx.newStreamAggrConfig()
sasNew, err := newStreamAggrConfigWithOpts(pushFunc, path, opts)
if err != nil {
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reloads_errors_total{path=%q}`, path)).Inc()
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, path)).Set(0)
logger.Errorf("cannot reload -remoteWrite.streamAggr.config=%q; continue using the previously loaded config; error: %s", path, err)
logger.Errorf("cannot reload stream aggregation config at %q; continue using the previously loaded config; error: %s", path, err)
return
}
sas := rwctx.sas.Load()
var sas *streamaggr.Aggregators
if idx < 0 {
sas = sasGlobal.Load()
} else {
sas = rwctxs[idx].sas.Load()
}
if !sasNew.Equal(sas) {
sasOld := rwctx.sas.Swap(sasNew)
var sasOld *streamaggr.Aggregators
if idx < 0 {
sasOld = sasGlobal.Swap(sasNew)
} else {
sasOld = rwctxs[idx].sas.Swap(sasNew)
}
sasOld.MustStop()
logger.Infof("successfully reloaded -remoteWrite.streamAggr.config=%q", path)
logger.Infof("successfully reloaded stream aggregation configs at %q", path)
} else {
sasNew.MustStop()
logger.Infof("-remoteWrite.streamAggr.config=%q wasn't changed since the last reload", path)
logger.Infof("successfully reloaded stream aggregation configs at %q", path)
}
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_successful{path=%q}`, path)).Set(1)
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_streamaggr_config_reload_success_timestamp_seconds{path=%q}`, path)).Set(fasttime.UnixTimestamp())
}
func newStreamAggrConfigGlobal() (*streamaggr.Aggregators, error) {
path := *streamAggrGlobalConfig
if path == "" {
return nil, nil
func getStreamAggrOpts(idx int) (string, *streamaggr.Options) {
if idx < 0 {
return *streamAggrGlobalConfig, &streamaggr.Options{
DedupInterval: streamAggrGlobalDedupInterval.Duration(),
DropInputLabels: *streamAggrGlobalDropInputLabels,
IgnoreOldSamples: *streamAggrGlobalIgnoreOldSamples,
IgnoreFirstIntervals: *streamAggrGlobalIgnoreFirstIntervals,
}
}
opts := &streamaggr.Options{
DedupInterval: streamAggrGlobalDedupInterval.Duration(),
DropInputLabels: *streamAggrGlobalDropInputLabels,
IgnoreOldSamples: *streamAggrGlobalIgnoreOldSamples,
IgnoreFirstIntervals: *streamAggrGlobalIgnoreFirstIntervals,
KeepInput: *streamAggrGlobalKeepInput,
}
sas, err := streamaggr.LoadFromFile(path, pushToRemoteStoragesTrackDropped, opts, "global")
if err != nil {
return nil, fmt.Errorf("cannot load -streamAggr.config=%q: %w", *streamAggrGlobalConfig, err)
}
return sas, nil
}
func (rwctx *remoteWriteCtx) newStreamAggrConfig() (*streamaggr.Aggregators, error) {
return newStreamAggrConfigPerURL(rwctx.idx, rwctx.pushInternalTrackDropped)
}
func newStreamAggrConfigPerURL(idx int, pushFunc streamaggr.PushFunc) (*streamaggr.Aggregators, error) {
path := streamAggrConfig.GetOptionalArg(idx)
if path == "" {
return nil, nil
}
alias := fmt.Sprintf("%d:secret-url", idx+1)
if *showRemoteWriteURL {
alias = fmt.Sprintf("%d:%s", idx+1, remoteWriteURLs.GetOptionalArg(idx))
}
opts := &streamaggr.Options{
opts := streamaggr.Options{
DedupInterval: streamAggrDedupInterval.GetOptionalArg(idx),
DropInputLabels: *streamAggrDropInputLabels,
IgnoreOldSamples: streamAggrIgnoreOldSamples.GetOptionalArg(idx),
IgnoreFirstIntervals: streamAggrIgnoreFirstIntervals.GetOptionalArg(idx),
KeepInput: streamAggrKeepInput.GetOptionalArg(idx),
IgnoreFirstIntervals: *streamAggrIgnoreFirstIntervals,
}
sas, err := streamaggr.LoadFromFile(path, pushFunc, opts, alias)
if err != nil {
return nil, fmt.Errorf("cannot load -remoteWrite.streamAggr.config=%q: %w", path, err)
if len(*streamAggrConfig) == 0 {
return "", &opts
}
return sas, nil
return streamAggrConfig.GetOptionalArg(idx), &opts
}
func newStreamAggrConfigWithOpts(pushFunc streamaggr.PushFunc, path string, opts *streamaggr.Options) (*streamaggr.Aggregators, error) {
if len(path) == 0 {
// Skip empty stream aggregation config.
return nil, nil
}
return streamaggr.LoadFromFile(path, pushFunc, opts)
}
func newStreamAggrConfig(idx int, pushFunc streamaggr.PushFunc) (*streamaggr.Aggregators, error) {
path, opts := getStreamAggrOpts(idx)
return newStreamAggrConfigWithOpts(pushFunc, path, opts)
}

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

@@ -12,12 +12,11 @@ import (
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/netutil"
)
var (
addr = flag.String("datasource.url", "", "Datasource compatible with Prometheus HTTP API. It can be single node VictoriaMetrics or vmselect endpoint. Required parameter. "+
"Supports address in the form of IP address with a port (e.g., http://127.0.0.1:8428) or DNS SRV record. "+
"Supports address in the form of IP address with a port (e.g., 127.0.0.1:8428) or DNS SRV record. "+
"See also -remoteRead.disablePathAppend and -datasource.showURL")
appendTypePrefix = flag.Bool("datasource.appendTypePrefix", false, "Whether to add type prefix to -datasource.url based on the query type. Set to true if sending different query types to the vmselect URL.")
showDatasourceURL = flag.Bool("datasource.showURL", false, "Whether to avoid stripping sensitive information such as auth headers or passwords from URLs in log messages or UI and exported metrics. "+
@@ -99,9 +98,9 @@ func Init(extraParams url.Values) (QuerierBuilder, error) {
tr, err := httputils.Transport(*addr, *tlsCertFile, *tlsKeyFile, *tlsCAFile, *tlsServerName, *tlsInsecureSkipVerify)
if err != nil {
return nil, fmt.Errorf("failed to create transport for -datasource.url=%q: %w", *addr, err)
return nil, fmt.Errorf("failed to create transport: %w", err)
}
tr.DialContext = netutil.NewStatDialFunc("vmalert_datasource")
tr.DialContext = httputils.GetStatDialFunc("vmalert_datasource")
tr.DisableKeepAlives = *disableKeepAlive
tr.MaxIdleConnsPerHost = *maxIdleConnections
if tr.MaxIdleConns != 0 && tr.MaxIdleConns < tr.MaxIdleConnsPerHost {

View File

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

View File

@@ -13,7 +13,7 @@ func BenchmarkMetrics(b *testing.B) {
var pi promInstant
if err := pi.Unmarshal(payload); err != nil {
b.Fatal(err.Error())
b.Fatalf(err.Error())
}
b.Run("Instant", func(b *testing.B) {
for i := 0; i < b.N; i++ {

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

View File

@@ -3,7 +3,6 @@ package notifier
import (
"context"
"encoding/json"
"fmt"
"net/http"
"net/http/httptest"
"strconv"
@@ -17,10 +16,10 @@ func TestAlertManager_Addr(t *testing.T) {
const addr = "http://localhost"
am, err := NewAlertManager(addr, nil, promauth.HTTPClientConfig{}, nil, 0)
if err != nil {
t.Fatalf("unexpected error: %s", err)
t.Errorf("unexpected error: %s", err)
}
if am.Addr() != addr {
t.Fatalf("expected to have %q; got %q", addr, am.Addr())
t.Errorf("expected to have %q; got %q", addr, am.Addr())
}
}
@@ -29,29 +28,27 @@ func TestAlertManager_Send(t *testing.T) {
const headerKey, headerValue = "TenantID", "foo"
mux := http.NewServeMux()
mux.HandleFunc("/", func(_ http.ResponseWriter, _ *http.Request) {
t.Fatalf("should not be called")
t.Errorf("should not be called")
})
c := -1
mux.HandleFunc(alertManagerPath, func(w http.ResponseWriter, r *http.Request) {
user, pass, ok := r.BasicAuth()
if !ok {
t.Fatalf("unauthorized request")
t.Errorf("unauthorized request")
}
if user != baUser || pass != baPass {
t.Fatalf("wrong creds %q:%q; expected %q:%q", user, pass, baUser, baPass)
t.Errorf("wrong creds %q:%q; expected %q:%q",
user, pass, baUser, baPass)
}
c++
if r.Method != http.MethodPost {
t.Fatalf("expected POST method got %s", r.Method)
t.Errorf("expected POST method got %s", r.Method)
}
switch c {
case 0:
conn, _, _ := w.(http.Hijacker).Hijack()
_ = conn.Close()
case 1:
if r.Header.Get(headerKey) != headerValue {
t.Fatalf("expected header %q to be set to %q; got %q instead", headerKey, headerValue, r.Header.Get(headerKey))
}
w.WriteHeader(500)
case 2:
var a []struct {
@@ -62,26 +59,25 @@ func TestAlertManager_Send(t *testing.T) {
GeneratorURL string `json:"generatorURL"`
}
if err := json.NewDecoder(r.Body).Decode(&a); err != nil {
t.Fatalf("can not unmarshal data into alert %s", err)
t.Errorf("can not unmarshal data into alert %s", err)
t.FailNow()
}
if len(a) != 1 {
t.Fatalf("expected 1 alert in array got %d", len(a))
t.Errorf("expected 1 alert in array got %d", len(a))
}
if a[0].GeneratorURL != "0/0" {
t.Fatalf("expected 0/0 as generatorURL got %s", a[0].GeneratorURL)
t.Errorf("expected 0/0 as generatorURL got %s", a[0].GeneratorURL)
}
if a[0].StartsAt.IsZero() {
t.Fatalf("expected non-zero start time")
t.Errorf("expected non-zero start time")
}
if a[0].EndAt.IsZero() {
t.Fatalf("expected non-zero end time")
}
if r.Header.Get(headerKey) != "bar" {
t.Fatalf("expected header %q to be set to %q; got %q instead", headerKey, headerValue, r.Header.Get(headerKey))
t.Errorf("expected non-zero end time")
}
case 3:
if r.Header.Get(headerKey) != headerValue {
t.Fatalf("expected header %q to be set to %q; got %q instead", headerKey, headerValue, r.Header.Get(headerKey))
t.Errorf("expected header %q to be set to %q; got %q instead",
headerKey, headerValue, r.Header.Get(headerKey))
}
}
})
@@ -93,19 +89,18 @@ func TestAlertManager_Send(t *testing.T) {
Username: baUser,
Password: promauth.NewSecret(baPass),
},
Headers: []string{fmt.Sprintf("%s:%s", headerKey, headerValue)},
}
am, err := NewAlertManager(srv.URL+alertManagerPath, func(alert Alert) string {
return strconv.FormatUint(alert.GroupID, 10) + "/" + strconv.FormatUint(alert.ID, 10)
}, aCfg, nil, 0)
if err != nil {
t.Fatalf("unexpected error: %s", err)
t.Errorf("unexpected error: %s", err)
}
if err := am.Send(context.Background(), []Alert{{}, {}}, nil); err == nil {
t.Fatalf("expected connection error got nil")
t.Error("expected connection error got nil")
}
if err := am.Send(context.Background(), []Alert{}, nil); err == nil {
t.Fatalf("expected wrong http code error got nil")
t.Error("expected wrong http code error got nil")
}
if err := am.Send(context.Background(), []Alert{{
GroupID: 0,
@@ -113,13 +108,13 @@ func TestAlertManager_Send(t *testing.T) {
Start: time.Now().UTC(),
End: time.Now().UTC(),
Annotations: map[string]string{"a": "b", "c": "d", "e": "f"},
}}, map[string]string{headerKey: "bar"}); err != nil {
t.Fatalf("unexpected error %s", err)
}}, nil); err != nil {
t.Errorf("unexpected error %s", err)
}
if c != 2 {
t.Fatalf("expected 2 calls(count from zero) to server got %d", c)
t.Errorf("expected 2 calls(count from zero) to server got %d", c)
}
if err := am.Send(context.Background(), nil, map[string]string{headerKey: headerValue}); err != nil {
t.Fatalf("unexpected error %s", err)
t.Errorf("unexpected error %s", err)
}
}

View File

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

View File

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

View File

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

View File

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

View File

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

View File

@@ -10,14 +10,13 @@ import (
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/utils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/netutil"
)
var (
addr = flag.String("remoteRead.url", "", "Optional URL to datasource compatible with Prometheus HTTP API. It can be single node VictoriaMetrics or vmselect."+
"Remote read is used to restore alerts state."+
"This configuration makes sense only if `vmalert` was configured with `remoteWrite.url` before and has been successfully persisted its state. "+
"Supports address in the form of IP address with a port (e.g., http://127.0.0.1:8428) or DNS SRV record. "+
"Supports address in the form of IP address with a port (e.g., 127.0.0.1:8428) or DNS SRV record. "+
"See also '-remoteRead.disablePathAppend', '-remoteRead.showURL'.")
showRemoteReadURL = flag.Bool("remoteRead.showURL", false, "Whether to show -remoteRead.url in the exported metrics. "+
@@ -68,10 +67,10 @@ func Init() (datasource.QuerierBuilder, error) {
}
tr, err := httputils.Transport(*addr, *tlsCertFile, *tlsKeyFile, *tlsCAFile, *tlsServerName, *tlsInsecureSkipVerify)
if err != nil {
return nil, fmt.Errorf("failed to create transport for -remoteRead.url=%q: %w", *addr, err)
return nil, fmt.Errorf("failed to create transport: %w", err)
}
tr.IdleConnTimeout = *idleConnectionTimeout
tr.DialContext = netutil.NewStatDialFunc("vmalert_remoteread")
tr.DialContext = httputils.GetStatDialFunc("vmalert_remoteread")
endpointParams, err := flagutil.ParseJSONMap(*oauth2EndpointParams)
if err != nil {

View File

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

View File

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

View File

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

View File

@@ -9,13 +9,12 @@ import (
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/utils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputils"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/netutil"
)
var (
addr = flag.String("remoteWrite.url", "", "Optional URL to VictoriaMetrics or vminsert where to persist alerts state "+
"and recording rules results in form of timeseries. "+
"Supports address in the form of IP address with a port (e.g., http://127.0.0.1:8428) or DNS SRV record. "+
"Supports address in the form of IP address with a port (e.g., 127.0.0.1:8428) or DNS SRV record. "+
"For example, if -remoteWrite.url=http://127.0.0.1:8428 is specified, "+
"then the alerts state will be written to http://127.0.0.1:8428/api/v1/write . See also -remoteWrite.disablePathAppend, '-remoteWrite.showURL'.")
showRemoteWriteURL = flag.Bool("remoteWrite.showURL", false, "Whether to show -remoteWrite.url in the exported metrics. "+
@@ -72,10 +71,10 @@ func Init(ctx context.Context) (*Client, error) {
t, err := httputils.Transport(*addr, *tlsCertFile, *tlsKeyFile, *tlsCAFile, *tlsServerName, *tlsInsecureSkipVerify)
if err != nil {
return nil, fmt.Errorf("failed to create transport for -remoteWrite.url=%q: %w", *addr, err)
return nil, fmt.Errorf("failed to create transport: %w", err)
}
t.IdleConnTimeout = *idleConnectionTimeout
t.DialContext = netutil.NewStatDialFunc("vmalert_remotewrite")
t.DialContext = httputils.GetStatDialFunc("vmalert_remotewrite")
endpointParams, err := flagutil.ParseJSONMap(*oauth2EndpointParams)
if err != nil {

View File

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

View File

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

View File

@@ -183,7 +183,7 @@ func (ar *AlertingRule) GetAlert(id uint64) *notifier.Alert {
return ar.alerts[id]
}
func (ar *AlertingRule) logDebugf(at time.Time, a *notifier.Alert, format string, args ...any) {
func (ar *AlertingRule) logDebugf(at time.Time, a *notifier.Alert, format string, args ...interface{}) {
if !ar.Debug {
return
}
@@ -441,6 +441,9 @@ func (ar *AlertingRule) exec(ctx context.Context, ts time.Time, limit int) ([]pr
}
a.Value = m.Values[0]
a.Annotations = annotations
if err != nil {
return nil, err
}
a.KeepFiringSince = time.Time{}
continue
}

File diff suppressed because it is too large Load Diff

View File

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

View File

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

View File

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

View File

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

View File

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

View File

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

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