mirror of
https://github.com/VictoriaMetrics/VictoriaMetrics.git
synced 2026-05-21 10:46:35 +03:00
Compare commits
1 Commits
docs/trace
...
issue-8584
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
c188d5cf08 |
8
.github/ISSUE_TEMPLATE/bug_report.yml
vendored
8
.github/ISSUE_TEMPLATE/bug_report.yml
vendored
@@ -5,10 +5,10 @@ body:
|
||||
- type: markdown
|
||||
attributes:
|
||||
value: |
|
||||
Before filling a bug report it would be great to [upgrade](https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#how-to-upgrade)
|
||||
Before filling a bug report it would be great to [upgrade](https://docs.victoriametrics.com/#how-to-upgrade)
|
||||
to [the latest available release](https://github.com/VictoriaMetrics/VictoriaMetrics/releases/latest)
|
||||
and verify whether the bug is reproducible there.
|
||||
It's also recommended to read the [troubleshooting docs](https://docs.victoriametrics.com/victoriametrics/troubleshooting/) first.
|
||||
It's also recommended to read the [troubleshooting docs](https://docs.victoriametrics.com/troubleshooting/) first.
|
||||
- type: textarea
|
||||
id: describe-the-bug
|
||||
attributes:
|
||||
@@ -64,8 +64,8 @@ body:
|
||||
* [Grafana dashboard for VictoriaMetrics cluster](https://grafana.com/grafana/dashboards/11176)
|
||||
|
||||
See how to setup monitoring here:
|
||||
* [monitoring for single-node VictoriaMetrics](https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#monitoring)
|
||||
* [monitoring for VictoriaMetrics cluster](https://docs.victoriametrics.com/victoriametrics/cluster-victoriametrics/#monitoring)
|
||||
* [monitoring for single-node VictoriaMetrics](https://docs.victoriametrics.com/#monitoring)
|
||||
* [monitoring for VictoriaMetrics cluster](https://docs.victoriametrics.com/cluster-victoriametrics/#monitoring)
|
||||
validations:
|
||||
required: false
|
||||
- type: textarea
|
||||
|
||||
6
.github/ISSUE_TEMPLATE/question.yml
vendored
6
.github/ISSUE_TEMPLATE/question.yml
vendored
@@ -24,9 +24,9 @@ body:
|
||||
label: Troubleshooting docs
|
||||
description: I am familiar with the following troubleshooting docs
|
||||
options:
|
||||
- label: General - https://docs.victoriametrics.com/victoriametrics/troubleshooting/
|
||||
- label: General - https://docs.victoriametrics.com/troubleshooting/
|
||||
required: false
|
||||
- label: vmagent - https://docs.victoriametrics.com/victoriametrics/vmagent/#troubleshooting
|
||||
- label: vmagent - https://docs.victoriametrics.com/vmagent/#troubleshooting
|
||||
required: false
|
||||
- label: vmalert - https://docs.victoriametrics.com/victoriametrics/vmalert/#troubleshooting
|
||||
- label: vmalert - https://docs.victoriametrics.com/vmalert/#troubleshooting
|
||||
required: false
|
||||
|
||||
2
.github/pull_request_template.md
vendored
2
.github/pull_request_template.md
vendored
@@ -6,4 +6,4 @@ Please provide a brief description of the changes you made. Be as specific as po
|
||||
|
||||
The following checks are **mandatory**:
|
||||
|
||||
- [ ] My change adheres to [VictoriaMetrics contributing guidelines](https://docs.victoriametrics.com/victoriametrics/contributing/#pull-request-checklist).
|
||||
- [ ] My change adheres [VictoriaMetrics contributing guidelines](https://docs.victoriametrics.com/contributing/).
|
||||
|
||||
1
.gitignore
vendored
1
.gitignore
vendored
@@ -27,4 +27,3 @@ _site
|
||||
coverage.txt
|
||||
cspell.json
|
||||
*~
|
||||
deployment/docker/provisioning/plugins/
|
||||
|
||||
@@ -1 +1 @@
|
||||
The document has been moved [here](https://docs.victoriametrics.com/victoriametrics/contributing/).
|
||||
The document has been moved [here](https://docs.victoriametrics.com/contributing/).
|
||||
|
||||
50
Makefile
50
Makefile
@@ -5,6 +5,7 @@ MAKE_PARALLEL := $(MAKE) -j $(MAKE_CONCURRENCY)
|
||||
DATEINFO_TAG ?= $(shell date -u +'%Y%m%d-%H%M%S')
|
||||
BUILDINFO_TAG ?= $(shell echo $$(git describe --long --all | tr '/' '-')$$( \
|
||||
git diff-index --quiet HEAD -- || echo '-dirty-'$$(git diff-index -u HEAD | openssl sha1 | cut -d' ' -f2 | cut -c 1-8)))
|
||||
LATEST_TAG ?= latest
|
||||
|
||||
PKG_TAG ?= $(shell git tag -l --points-at HEAD)
|
||||
ifeq ($(PKG_TAG),)
|
||||
@@ -195,31 +196,12 @@ vmutils-crossbuild: \
|
||||
vmutils-openbsd-amd64 \
|
||||
vmutils-windows-amd64
|
||||
|
||||
publish-latest:
|
||||
PKG_TAG=$(TAG) APP_NAME=victoria-metrics $(MAKE) publish-via-docker-latest && \
|
||||
PKG_TAG=$(TAG) APP_NAME=vmagent $(MAKE) publish-via-docker-latest && \
|
||||
PKG_TAG=$(TAG) APP_NAME=vmalert $(MAKE) publish-via-docker-latest && \
|
||||
PKG_TAG=$(TAG) APP_NAME=vmalert-tool $(MAKE) publish-via-docker-latest && \
|
||||
PKG_TAG=$(TAG) APP_NAME=vmauth $(MAKE) publish-via-docker-latest && \
|
||||
PKG_TAG=$(TAG) APP_NAME=vmbackup $(MAKE) publish-via-docker-latest && \
|
||||
PKG_TAG=$(TAG) APP_NAME=vmrestore $(MAKE) publish-via-docker-latest && \
|
||||
PKG_TAG=$(TAG) APP_NAME=vmctl $(MAKE) publish-via-docker-latest && \
|
||||
PKG_TAG=$(TAG)-cluster APP_NAME=vminsert $(MAKE) publish-via-docker-latest && \
|
||||
PKG_TAG=$(TAG)-cluster APP_NAME=vmselect $(MAKE) publish-via-docker-latest && \
|
||||
PKG_TAG=$(TAG)-cluster APP_NAME=vmstorage $(MAKE) publish-via-docker-latest && \
|
||||
PKG_TAG=$(TAG)-enterprise APP_NAME=vmgateway $(MAKE) publish-via-docker-latest
|
||||
PKG_TAG=$(TAG)-enterprise APP_NAME=vmbackupmanager $(MAKE) publish-via-docker-latest
|
||||
|
||||
publish-victoria-logs-latest:
|
||||
PKG_TAG=$(TAG) APP_NAME=victoria-logs $(MAKE) publish-via-docker-latest
|
||||
PKG_TAG=$(TAG) APP_NAME=vlogscli $(MAKE) publish-via-docker-latest
|
||||
|
||||
publish-release:
|
||||
rm -rf bin/*
|
||||
git checkout $(TAG) && $(MAKE) release && $(MAKE) publish && \
|
||||
git checkout $(TAG)-cluster && $(MAKE) release && $(MAKE) publish && \
|
||||
git checkout $(TAG)-enterprise && $(MAKE) release && $(MAKE) publish && \
|
||||
git checkout $(TAG)-enterprise-cluster && $(MAKE) release && $(MAKE) publish
|
||||
git checkout $(TAG) && $(MAKE) release && LATEST_TAG=stable $(MAKE) publish && \
|
||||
git checkout $(TAG)-cluster && $(MAKE) release && LATEST_TAG=cluster-stable $(MAKE) publish && \
|
||||
git checkout $(TAG)-enterprise && $(MAKE) release && LATEST_TAG=enterprise-stable $(MAKE) publish && \
|
||||
git checkout $(TAG)-enterprise-cluster && $(MAKE) release && LATEST_TAG=enterprise-cluster-stable $(MAKE) publish
|
||||
|
||||
release:
|
||||
$(MAKE_PARALLEL) \
|
||||
@@ -522,7 +504,7 @@ fmt:
|
||||
gofmt -l -w -s ./apptest
|
||||
|
||||
vet:
|
||||
GOEXPERIMENT=synctest go vet ./lib/...
|
||||
go vet ./lib/...
|
||||
go vet ./app/...
|
||||
go vet ./apptest/...
|
||||
|
||||
@@ -531,35 +513,35 @@ check-all: fmt vet golangci-lint govulncheck
|
||||
clean-checkers: remove-golangci-lint remove-govulncheck
|
||||
|
||||
test:
|
||||
GOEXPERIMENT=synctest go test ./lib/... ./app/...
|
||||
go test ./lib/... ./app/...
|
||||
|
||||
test-race:
|
||||
GOEXPERIMENT=synctest go test -race ./lib/... ./app/...
|
||||
go test -race ./lib/... ./app/...
|
||||
|
||||
test-pure:
|
||||
GOEXPERIMENT=synctest CGO_ENABLED=0 go test ./lib/... ./app/...
|
||||
CGO_ENABLED=0 go test ./lib/... ./app/...
|
||||
|
||||
test-full:
|
||||
GOEXPERIMENT=synctest go test -coverprofile=coverage.txt -covermode=atomic ./lib/... ./app/...
|
||||
go test -coverprofile=coverage.txt -covermode=atomic ./lib/... ./app/...
|
||||
|
||||
test-full-386:
|
||||
GOEXPERIMENT=synctest GOARCH=386 go test -coverprofile=coverage.txt -covermode=atomic ./lib/... ./app/...
|
||||
GOARCH=386 go test -coverprofile=coverage.txt -covermode=atomic ./lib/... ./app/...
|
||||
|
||||
integration-test: victoria-metrics vmagent vmalert vmauth vmctl vmbackup vmrestore victoria-logs
|
||||
integration-test: victoria-metrics vmagent vmalert vmauth
|
||||
go test ./apptest/... -skip="^TestCluster.*"
|
||||
|
||||
benchmark:
|
||||
GOEXPERIMENT=synctest go test -bench=. ./lib/...
|
||||
go test -bench=. ./lib/...
|
||||
go test -bench=. ./app/...
|
||||
|
||||
benchmark-pure:
|
||||
GOEXPERIMENT=synctest CGO_ENABLED=0 go test -bench=. ./lib/...
|
||||
CGO_ENABLED=0 go test -bench=. ./lib/...
|
||||
CGO_ENABLED=0 go test -bench=. ./app/...
|
||||
|
||||
vendor-update:
|
||||
go get -u ./lib/...
|
||||
go get -u ./app/...
|
||||
go mod tidy -compat=1.24
|
||||
go mod tidy -compat=1.23
|
||||
go mod vendor
|
||||
|
||||
app-local:
|
||||
@@ -582,7 +564,7 @@ install-qtc:
|
||||
|
||||
|
||||
golangci-lint: install-golangci-lint
|
||||
GOEXPERIMENT=synctest golangci-lint run
|
||||
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 v1.64.7
|
||||
|
||||
44
README.md
44
README.md
@@ -11,9 +11,9 @@
|
||||

|
||||
|
||||
<picture>
|
||||
<source srcset="docs/victoriametrics/logo_white.webp" media="(prefers-color-scheme: dark)">
|
||||
<source srcset="docs/victoriametrics/logo.webp" media="(prefers-color-scheme: light)">
|
||||
<img src="docs/victoriametrics/logo.webp" width="300" alt="VictoriaMetrics logo">
|
||||
<source srcset="docs/logo_white.webp" media="(prefers-color-scheme: dark)">
|
||||
<source srcset="docs/logo.webp" media="(prefers-color-scheme: light)">
|
||||
<img src="docs/logo.webp" width="300" alt="VictoriaMetrics logo">
|
||||
</picture>
|
||||
|
||||
VictoriaMetrics is a fast, cost-saving, and scalable solution for monitoring and managing time series data. It delivers high performance and reliability, making it an ideal choice for businesses of all sizes.
|
||||
@@ -21,10 +21,10 @@ VictoriaMetrics is a fast, cost-saving, and scalable solution for monitoring and
|
||||
Here are some resources and information about VictoriaMetrics:
|
||||
|
||||
- Documentation: [docs.victoriametrics.com](https://docs.victoriametrics.com)
|
||||
- Case studies: [Grammarly, Roblox, Wix,...](https://docs.victoriametrics.com/victoriametrics/casestudies/).
|
||||
- Case studies: [Grammarly, Roblox, Wix,...](https://docs.victoriametrics.com/casestudies/).
|
||||
- Available: [Binary releases](https://github.com/VictoriaMetrics/VictoriaMetrics/releases/latest), docker images [Docker Hub](https://hub.docker.com/r/victoriametrics/victoria-metrics/) and [Quay](https://quay.io/repository/victoriametrics/victoria-metrics), [Source code](https://github.com/VictoriaMetrics/VictoriaMetrics)
|
||||
- Deployment types: [Single-node version](https://docs.victoriametrics.com/), [Cluster version](https://docs.victoriametrics.com/victoriametrics/cluster-victoriametrics/), and [Enterprise version](https://docs.victoriametrics.com/victoriametrics/enterprise/)
|
||||
- Changelog: [CHANGELOG](https://docs.victoriametrics.com/victoriametrics/changelog/), and [How to upgrade](https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#how-to-upgrade-victoriametrics)
|
||||
- Deployment types: [Single-node version](https://docs.victoriametrics.com/), [Cluster version](https://docs.victoriametrics.com/cluster-victoriametrics/), and [Enterprise version](https://docs.victoriametrics.com/enterprise/)
|
||||
- Changelog: [CHANGELOG](https://docs.victoriametrics.com/changelog/), and [How to upgrade](https://docs.victoriametrics.com/#how-to-upgrade-victoriametrics)
|
||||
- Community: [Slack](https://slack.victoriametrics.com/), [X (Twitter)](https://x.com/VictoriaMetrics), [LinkedIn](https://www.linkedin.com/company/victoriametrics/), [YouTube](https://www.youtube.com/@VictoriaMetrics)
|
||||
|
||||
Yes, we open-source both the single-node VictoriaMetrics and the cluster version.
|
||||
@@ -35,22 +35,22 @@ VictoriaMetrics is optimized for timeseries data, even when old time series are
|
||||
|
||||
* **Long-term storage for Prometheus** or as a drop-in replacement for Prometheus and Graphite in Grafana.
|
||||
* **Powerful stream aggregation**: Can be used as a StatsD alternative.
|
||||
* **Ideal for big data**: Works well with large amounts of time series data from APM, Kubernetes, IoT sensors, connected cars, industrial telemetry, financial data and various [Enterprise workloads](https://docs.victoriametrics.com/victoriametrics/enterprise/).
|
||||
* **Ideal for big data**: Works well with large amounts of time series data from APM, Kubernetes, IoT sensors, connected cars, industrial telemetry, financial data and various [Enterprise workloads](https://docs.victoriametrics.com/enterprise/).
|
||||
* **Query language**: Supports both PromQL and the more performant MetricsQL.
|
||||
* **Easy to setup**: No dependencies, single [small binary](https://medium.com/@valyala/stripping-dependency-bloat-in-victoriametrics-docker-image-983fb5912b0d), configuration through command-line flags, but the default is also fine-tuned; backup and restore with [instant snapshots](https://medium.com/@valyala/how-victoriametrics-makes-instant-snapshots-for-multi-terabyte-time-series-data-e1f3fb0e0282).
|
||||
* **Global query view**: Multiple Prometheus instances or any other data sources may ingest data into VictoriaMetrics and queried via a single query.
|
||||
* **Various Protocols**: Support metric scraping, ingestion and backfilling in various protocol.
|
||||
* [Prometheus exporters](https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#how-to-scrape-prometheus-exporters-such-as-node-exporter), [Prometheus remote write API](https://docs.victoriametrics.com/victoriametrics/integrations/prometheus/), [Prometheus exposition format](https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#how-to-import-data-in-prometheus-exposition-format).
|
||||
* [InfluxDB line protocol](https://docs.victoriametrics.com/victoriametrics/integrations/influxdb/) over HTTP, TCP and UDP.
|
||||
* [Graphite plaintext protocol](https://docs.victoriametrics.com/victoriametrics/integrations/graphite/#ingesting) with [tags](https://graphite.readthedocs.io/en/latest/tags.html#carbon).
|
||||
* [OpenTSDB put message](https://docs.victoriametrics.com/victoriametrics/integrations/opentsdb/#sending-data-via-telnet).
|
||||
* [HTTP OpenTSDB /api/put requests](https://docs.victoriametrics.com/victoriametrics/integrations/opentsdb/#sending-data-via-http).
|
||||
* [JSON line format](https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#how-to-import-data-in-json-line-format).
|
||||
* [Arbitrary CSV data](https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#how-to-import-csv-data).
|
||||
* [Native binary format](https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#how-to-import-data-in-native-format).
|
||||
* [DataDog agent or DogStatsD](https://docs.victoriametrics.com/victoriametrics/integrations/datadog/).
|
||||
* [NewRelic infrastructure agent](https://docs.victoriametrics.com/victoriametrics/integrations/newrelic/#sending-data-from-agent).
|
||||
* [OpenTelemetry metrics format](https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#sending-data-via-opentelemetry).
|
||||
* [Prometheus exporters](https://docs.victoriametrics.com/#how-to-scrape-prometheus-exporters-such-as-node-exporter), [Prometheus remote write API](https://docs.victoriametrics.com/#prometheus-setup), [Prometheus exposition format](https://docs.victoriametrics.com/#how-to-import-data-in-prometheus-exposition-format).
|
||||
* [InfluxDB line protocol](https://docs.victoriametrics.com/#how-to-send-data-from-influxdb-compatible-agents-such-as-telegraf) over HTTP, TCP and UDP.
|
||||
* [Graphite plaintext protocol](https://docs.victoriametrics.com/#how-to-send-data-from-graphite-compatible-agents-such-as-statsd) with [tags](https://graphite.readthedocs.io/en/latest/tags.html#carbon).
|
||||
* [OpenTSDB put message](https://docs.victoriametrics.com/#sending-data-via-telnet-put-protocol).
|
||||
* [HTTP OpenTSDB /api/put requests](https://docs.victoriametrics.com/#sending-opentsdb-data-via-http-apiput-requests).
|
||||
* [JSON line format](https://docs.victoriametrics.com/#how-to-import-data-in-json-line-format).
|
||||
* [Arbitrary CSV data](https://docs.victoriametrics.com/#how-to-import-csv-data).
|
||||
* [Native binary format](https://docs.victoriametrics.com/#how-to-import-data-in-native-format).
|
||||
* [DataDog agent or DogStatsD](https://docs.victoriametrics.com/#how-to-send-data-from-datadog-agent).
|
||||
* [NewRelic infrastructure agent](https://docs.victoriametrics.com/#how-to-send-data-from-newrelic-agent).
|
||||
* [OpenTelemetry metrics format](https://docs.victoriametrics.com/#sending-data-via-opentelemetry).
|
||||
* **NFS-based storages**: Supports storing data on NFS-based storages such as Amazon EFS, Google Filestore.
|
||||
* And many other features such as metrics relabeling, cardinality limiter, etc.
|
||||
|
||||
@@ -62,9 +62,9 @@ In addition, the Enterprise version includes extra features:
|
||||
- **Backup automation**: Automates regular backup procedures.
|
||||
- **Multiple retentions**: Reducing storage costs by specifying different retentions for different datasets.
|
||||
- **Downsampling**: Reducing storage costs and increasing performance for queries over historical data.
|
||||
- **Stable releases** with long-term support lines ([LTS](https://docs.victoriametrics.com/victoriametrics/lts-releases/)).
|
||||
- **Stable releases** with long-term support lines ([LTS](https://docs.victoriametrics.com/lts-releases/)).
|
||||
- **Comprehensive support**: First-class consulting, feature requests and technical support provided by the core VictoriaMetrics dev team.
|
||||
- Many other features, which you can read about on [the Enterprise page](https://docs.victoriametrics.com/victoriametrics/enterprise/).
|
||||
- Many other features, which you can read about on [the Enterprise page](https://docs.victoriametrics.com/enterprise/).
|
||||
|
||||
[Contact us](mailto:info@victoriametrics.com) if you need enterprise support for VictoriaMetrics. Or you can request a free trial license [here](https://victoriametrics.com/products/enterprise/trial/), downloaded Enterprise binaries are available at [Github Releases](https://github.com/VictoriaMetrics/VictoriaMetrics/releases/latest).
|
||||
|
||||
@@ -77,7 +77,7 @@ Some good benchmarks VictoriaMetrics achieved:
|
||||
* **Minimal memory footprint**: handling millions of unique timeseries with [10x less RAM](https://medium.com/@valyala/insert-benchmarks-with-inch-influxdb-vs-victoriametrics-e31a41ae2893) than InfluxDB, up to [7x less RAM](https://valyala.medium.com/prometheus-vs-victoriametrics-benchmark-on-node-exporter-metrics-4ca29c75590f) than Prometheus, Thanos or Cortex.
|
||||
* **Highly scalable and performance** for [data ingestion](https://medium.com/@valyala/high-cardinality-tsdb-benchmarks-victoriametrics-vs-timescaledb-vs-influxdb-13e6ee64dd6b) and [querying](https://medium.com/@valyala/when-size-matters-benchmarking-victoriametrics-vs-timescale-and-influxdb-6035811952d4), [20x outperforms](https://medium.com/@valyala/insert-benchmarks-with-inch-influxdb-vs-victoriametrics-e31a41ae2893) InfluxDB and TimescaleDB.
|
||||
* **High data compression**: [70x more data points](https://medium.com/@valyala/when-size-matters-benchmarking-victoriametrics-vs-timescale-and-influxdb-6035811952d4) may be stored into limited storage than TimescaleDB, [7x less storage](https://valyala.medium.com/prometheus-vs-victoriametrics-benchmark-on-node-exporter-metrics-4ca29c75590f) space is required than Prometheus, Thanos or Cortex.
|
||||
* **Reducing storage costs**: [10x more effective](https://docs.victoriametrics.com/victoriametrics/casestudies/#grammarly) than Graphite according to the Grammarly case study.
|
||||
* **Reducing storage costs**: [10x more effective](https://docs.victoriametrics.com/casestudies/#grammarly) than Graphite according to the Grammarly case study.
|
||||
* **A single-node VictoriaMetrics** can replace medium-sized clusters built with competing solutions such as Thanos, M3DB, Cortex, InfluxDB or TimescaleDB. See [VictoriaMetrics vs Thanos](https://medium.com/@valyala/comparing-thanos-to-victoriametrics-cluster-b193bea1683), [Measuring vertical scalability](https://medium.com/@valyala/measuring-vertical-scalability-for-time-series-databases-in-google-cloud-92550d78d8ae), [Remote write storage wars - PromCon 2019](https://promcon.io/2019-munich/talks/remote-write-storage-wars/).
|
||||
* **Optimized for storage**: [Works well with high-latency IO](https://medium.com/@valyala/high-cardinality-tsdb-benchmarks-victoriametrics-vs-timescaledb-vs-influxdb-13e6ee64dd6b) and low IOPS (HDD and network storage in AWS, Google Cloud, Microsoft Azure, etc.).
|
||||
|
||||
@@ -93,7 +93,7 @@ Feel free asking any questions regarding VictoriaMetrics:
|
||||
* [Telegram-ru](https://t.me/VictoriaMetrics_ru1)
|
||||
* [Mastodon](https://mastodon.social/@victoriametrics/)
|
||||
|
||||
If you like VictoriaMetrics and want to contribute, then please [read these docs](https://docs.victoriametrics.com/victoriametrics/contributing/).
|
||||
If you like VictoriaMetrics and want to contribute, then please [read these docs](https://docs.victoriametrics.com/contributing/).
|
||||
|
||||
## VictoriaMetrics Logo
|
||||
|
||||
|
||||
@@ -6,9 +6,9 @@ The following versions of VictoriaMetrics receive regular security fixes:
|
||||
|
||||
| Version | Supported |
|
||||
|---------|--------------------|
|
||||
| [latest release](https://docs.victoriametrics.com/victoriametrics/changelog/) | :white_check_mark: |
|
||||
| v1.102.x [LTS line](https://docs.victoriametrics.com/victoriametrics/lts-releases/) | :white_check_mark: |
|
||||
| v1.110.x [LTS line](https://docs.victoriametrics.com/victoriametrics/lts-releases/) | :white_check_mark: |
|
||||
| [latest release](https://docs.victoriametrics.com/changelog/) | :white_check_mark: |
|
||||
| v1.102.x [LTS line](https://docs.victoriametrics.com/lts-releases/) | :white_check_mark: |
|
||||
| v1.110.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.
|
||||
|
||||
@@ -8,7 +8,6 @@ import (
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlselect"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/buildinfo"
|
||||
@@ -45,13 +44,9 @@ func main() {
|
||||
|
||||
vlstorage.Init()
|
||||
vlselect.Init()
|
||||
|
||||
insertutil.SetLogRowsStorage(&vlstorage.Storage{})
|
||||
vlinsert.Init()
|
||||
|
||||
go httpserver.Serve(listenAddrs, requestHandler, httpserver.ServeOptions{
|
||||
UseProxyProtocol: useProxyProtocol,
|
||||
})
|
||||
go httpserver.Serve(listenAddrs, useProxyProtocol, requestHandler)
|
||||
logger.Infof("started VictoriaLogs in %.3f seconds; see https://docs.victoriametrics.com/victorialogs/", time.Since(startTime).Seconds())
|
||||
|
||||
pushmetrics.Init()
|
||||
|
||||
@@ -32,7 +32,7 @@ var (
|
||||
"See https://www.haproxy.org/download/1.8/doc/proxy-protocol.txt . "+
|
||||
"With enabled proxy protocol http server cannot serve regular /metrics endpoint. Use -pushmetrics.url for metrics pushing")
|
||||
minScrapeInterval = flag.Duration("dedup.minScrapeInterval", 0, "Leave only the last sample in every time series per each discrete interval "+
|
||||
"equal to -dedup.minScrapeInterval > 0. See also -streamAggr.dedupInterval and https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#deduplication")
|
||||
"equal to -dedup.minScrapeInterval > 0. See also -streamAggr.dedupInterval and https://docs.victoriametrics.com/#deduplication")
|
||||
dryRun = flag.Bool("dryRun", false, "Whether to check config files without running VictoriaMetrics. The following config files are checked: "+
|
||||
"-promscrape.config, -relabelConfig and -streamAggr.config. Unknown config entries aren't allowed in -promscrape.config by default. "+
|
||||
"This can be changed with -promscrape.config.strictParse=false command-line flag")
|
||||
@@ -45,7 +45,7 @@ var (
|
||||
finalDedupScheduleInterval = flag.Duration("storage.finalDedupScheduleCheckInterval", time.Hour, "The interval for checking when final deduplication process should be started."+
|
||||
"Storage unconditionally adds 25% jitter to the interval value on each check evaluation."+
|
||||
" Changing the interval to the bigger values may delay downsampling, deduplication for historical data."+
|
||||
" See also https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#deduplication")
|
||||
" See also https://docs.victoriametrics.com/#deduplication")
|
||||
)
|
||||
|
||||
func main() {
|
||||
@@ -101,9 +101,7 @@ func main() {
|
||||
|
||||
startSelfScraper()
|
||||
|
||||
go httpserver.Serve(listenAddrs, requestHandler, httpserver.ServeOptions{
|
||||
UseProxyProtocol: useProxyProtocol,
|
||||
})
|
||||
go httpserver.Serve(listenAddrs, useProxyProtocol, requestHandler)
|
||||
logger.Infof("started VictoriaMetrics in %.3f seconds", time.Since(startTime).Seconds())
|
||||
|
||||
pushmetrics.Init()
|
||||
|
||||
619
app/victoria-metrics/main_test.go
Normal file
619
app/victoria-metrics/main_test.go
Normal file
@@ -0,0 +1,619 @@
|
||||
package main
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"encoding/json"
|
||||
"errors"
|
||||
"flag"
|
||||
"fmt"
|
||||
"io"
|
||||
"log"
|
||||
"math/rand"
|
||||
"net"
|
||||
"net/http"
|
||||
"os"
|
||||
"path/filepath"
|
||||
"reflect"
|
||||
"strconv"
|
||||
"strings"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
testutil "github.com/VictoriaMetrics/VictoriaMetrics/app/victoria-metrics/test"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vminsert"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmselect"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmselect/promql"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/fs"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
|
||||
)
|
||||
|
||||
const (
|
||||
testFixturesDir = "testdata"
|
||||
testStorageSuffix = "vm-test-storage"
|
||||
testHTTPListenAddr = ":7654"
|
||||
testStatsDListenAddr = ":2003"
|
||||
testOpenTSDBListenAddr = ":4242"
|
||||
testOpenTSDBHTTPListenAddr = ":4243"
|
||||
testLogLevel = "INFO"
|
||||
)
|
||||
|
||||
const (
|
||||
testReadHTTPPath = "http://127.0.0.1" + testHTTPListenAddr
|
||||
testWriteHTTPPath = "http://127.0.0.1" + testHTTPListenAddr + "/write"
|
||||
testOpenTSDBWriteHTTPPath = "http://127.0.0.1" + testOpenTSDBHTTPListenAddr + "/api/put"
|
||||
testPromWriteHTTPPath = "http://127.0.0.1" + testHTTPListenAddr + "/api/v1/write"
|
||||
testImportCSVWriteHTTPPath = "http://127.0.0.1" + testHTTPListenAddr + "/api/v1/import/csv"
|
||||
|
||||
testHealthHTTPPath = "http://127.0.0.1" + testHTTPListenAddr + "/health"
|
||||
)
|
||||
|
||||
const (
|
||||
testStorageInitTimeout = 10 * time.Second
|
||||
)
|
||||
|
||||
var (
|
||||
storagePath string
|
||||
insertionTime = time.Now().UTC()
|
||||
)
|
||||
|
||||
type test struct {
|
||||
Name string `json:"name"`
|
||||
Data []string `json:"data"`
|
||||
InsertQuery string `json:"insert_query"`
|
||||
Query []string `json:"query"`
|
||||
ResultMetrics []Metric `json:"result_metrics"`
|
||||
ResultSeries Series `json:"result_series"`
|
||||
ResultQuery Query `json:"result_query"`
|
||||
Issue string `json:"issue"`
|
||||
ExpectedResultLinesCount int `json:"expected_result_lines_count"`
|
||||
}
|
||||
|
||||
type Metric struct {
|
||||
Metric map[string]string `json:"metric"`
|
||||
Values []float64 `json:"values"`
|
||||
Timestamps []int64 `json:"timestamps"`
|
||||
}
|
||||
|
||||
func (r *Metric) UnmarshalJSON(b []byte) error {
|
||||
type plain Metric
|
||||
return json.Unmarshal(testutil.PopulateTimeTpl(b, insertionTime), (*plain)(r))
|
||||
}
|
||||
|
||||
type Series struct {
|
||||
Status string `json:"status"`
|
||||
Data []map[string]string `json:"data"`
|
||||
}
|
||||
|
||||
type Query struct {
|
||||
Status string `json:"status"`
|
||||
Data struct {
|
||||
ResultType string `json:"resultType"`
|
||||
Result json.RawMessage `json:"result"`
|
||||
} `json:"data"`
|
||||
}
|
||||
|
||||
const rtVector, rtMatrix = "vector", "matrix"
|
||||
|
||||
func (q *Query) metrics() ([]Metric, error) {
|
||||
switch q.Data.ResultType {
|
||||
case rtVector:
|
||||
var r QueryInstant
|
||||
if err := json.Unmarshal(q.Data.Result, &r.Result); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
return r.metrics()
|
||||
case rtMatrix:
|
||||
var r QueryRange
|
||||
if err := json.Unmarshal(q.Data.Result, &r.Result); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
return r.metrics()
|
||||
default:
|
||||
return nil, fmt.Errorf("unknown result type %q", q.Data.ResultType)
|
||||
}
|
||||
}
|
||||
|
||||
type QueryInstant struct {
|
||||
Result []struct {
|
||||
Labels map[string]string `json:"metric"`
|
||||
TV [2]any `json:"value"`
|
||||
} `json:"result"`
|
||||
}
|
||||
|
||||
func (q QueryInstant) metrics() ([]Metric, error) {
|
||||
result := make([]Metric, len(q.Result))
|
||||
for i, res := range q.Result {
|
||||
f, err := strconv.ParseFloat(res.TV[1].(string), 64)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("metric %v, unable to parse float64 from %s: %w", res, res.TV[1], err)
|
||||
}
|
||||
var m Metric
|
||||
m.Metric = res.Labels
|
||||
m.Timestamps = append(m.Timestamps, int64(res.TV[0].(float64)))
|
||||
m.Values = append(m.Values, f)
|
||||
result[i] = m
|
||||
}
|
||||
return result, nil
|
||||
}
|
||||
|
||||
type QueryRange struct {
|
||||
Result []struct {
|
||||
Metric map[string]string `json:"metric"`
|
||||
Values [][]any `json:"values"`
|
||||
} `json:"result"`
|
||||
}
|
||||
|
||||
func (q QueryRange) metrics() ([]Metric, error) {
|
||||
var result []Metric
|
||||
for i, res := range q.Result {
|
||||
var m Metric
|
||||
for _, tv := range res.Values {
|
||||
f, err := strconv.ParseFloat(tv[1].(string), 64)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("metric %v, unable to parse float64 from %s: %w", res, tv[1], err)
|
||||
}
|
||||
m.Values = append(m.Values, f)
|
||||
m.Timestamps = append(m.Timestamps, int64(tv[0].(float64)))
|
||||
}
|
||||
if len(m.Values) < 1 || len(m.Timestamps) < 1 {
|
||||
return nil, fmt.Errorf("metric %v contains no values", res)
|
||||
}
|
||||
m.Metric = q.Result[i].Metric
|
||||
result = append(result, m)
|
||||
}
|
||||
return result, nil
|
||||
}
|
||||
|
||||
func (q *Query) UnmarshalJSON(b []byte) error {
|
||||
type plain Query
|
||||
return json.Unmarshal(testutil.PopulateTimeTpl(b, insertionTime), (*plain)(q))
|
||||
}
|
||||
|
||||
func TestMain(m *testing.M) {
|
||||
setUp()
|
||||
code := m.Run()
|
||||
tearDown()
|
||||
os.Exit(code)
|
||||
}
|
||||
|
||||
func setUp() {
|
||||
storagePath = filepath.Join(os.TempDir(), testStorageSuffix)
|
||||
processFlags()
|
||||
logger.Init()
|
||||
vmstorage.Init(promql.ResetRollupResultCacheIfNeeded)
|
||||
vmselect.Init()
|
||||
vminsert.Init()
|
||||
go httpserver.Serve(*httpListenAddrs, useProxyProtocol, requestHandler)
|
||||
readyStorageCheckFunc := func() bool {
|
||||
resp, err := http.Get(testHealthHTTPPath)
|
||||
if err != nil {
|
||||
return false
|
||||
}
|
||||
_ = resp.Body.Close()
|
||||
return resp.StatusCode == 200
|
||||
}
|
||||
if err := waitFor(testStorageInitTimeout, readyStorageCheckFunc); err != nil {
|
||||
log.Fatalf("http server can't start for %s seconds, err %s", testStorageInitTimeout, err)
|
||||
}
|
||||
}
|
||||
|
||||
func processFlags() {
|
||||
flag.Parse()
|
||||
for _, fv := range []struct {
|
||||
flag string
|
||||
value string
|
||||
}{
|
||||
{flag: "storageDataPath", value: storagePath},
|
||||
{flag: "httpListenAddr", value: testHTTPListenAddr},
|
||||
{flag: "graphiteListenAddr", value: testStatsDListenAddr},
|
||||
{flag: "opentsdbListenAddr", value: testOpenTSDBListenAddr},
|
||||
{flag: "loggerLevel", value: testLogLevel},
|
||||
{flag: "opentsdbHTTPListenAddr", value: testOpenTSDBHTTPListenAddr},
|
||||
} {
|
||||
// panics if flag doesn't exist
|
||||
if err := flag.Lookup(fv.flag).Value.Set(fv.value); err != nil {
|
||||
log.Fatalf("unable to set %q with value %q, err: %v", fv.flag, fv.value, err)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func waitFor(timeout time.Duration, f func() bool) error {
|
||||
fraction := timeout / 10
|
||||
for i := fraction; i < timeout; i += fraction {
|
||||
if f() {
|
||||
return nil
|
||||
}
|
||||
time.Sleep(fraction)
|
||||
}
|
||||
return fmt.Errorf("timeout")
|
||||
}
|
||||
|
||||
func tearDown() {
|
||||
if err := httpserver.Stop(*httpListenAddrs); err != nil {
|
||||
log.Printf("cannot stop the webservice: %s", err)
|
||||
}
|
||||
vminsert.Stop()
|
||||
vmstorage.Stop()
|
||||
vmselect.Stop()
|
||||
fs.MustRemoveAll(storagePath)
|
||||
}
|
||||
|
||||
func TestWriteRead(t *testing.T) {
|
||||
t.Run("write", testWrite)
|
||||
time.Sleep(500 * time.Millisecond)
|
||||
vmstorage.Storage.DebugFlush()
|
||||
time.Sleep(1500 * time.Millisecond)
|
||||
t.Run("read", testRead)
|
||||
}
|
||||
|
||||
func testWrite(t *testing.T) {
|
||||
t.Run("prometheus", func(t *testing.T) {
|
||||
for _, test := range readIn("prometheus", insertionTime) {
|
||||
if test.Data == nil {
|
||||
continue
|
||||
}
|
||||
r := testutil.WriteRequest{}
|
||||
testData := strings.Join(test.Data, "\n")
|
||||
if err := json.Unmarshal([]byte(testData), &r.Timeseries); err != nil {
|
||||
panic(fmt.Errorf("BUG: cannot unmarshal TimeSeries: %s\ntest data\n%s", err, testData))
|
||||
}
|
||||
if n := len(r.Timeseries); n <= 0 {
|
||||
panic(fmt.Errorf("BUG: expecting non-empty Timeseries in test data:\n%s", testData))
|
||||
}
|
||||
data := testutil.Compress(r)
|
||||
httpWrite(t, testPromWriteHTTPPath, test.InsertQuery, bytes.NewBuffer(data))
|
||||
}
|
||||
})
|
||||
t.Run("csv", func(t *testing.T) {
|
||||
for _, test := range readIn("csv", insertionTime) {
|
||||
if test.Data == nil {
|
||||
continue
|
||||
}
|
||||
httpWrite(t, testImportCSVWriteHTTPPath, test.InsertQuery, bytes.NewBuffer([]byte(strings.Join(test.Data, "\n"))))
|
||||
}
|
||||
})
|
||||
|
||||
t.Run("influxdb", func(t *testing.T) {
|
||||
for _, x := range readIn("influxdb", insertionTime) {
|
||||
test := x
|
||||
t.Run(test.Name, func(t *testing.T) {
|
||||
t.Parallel()
|
||||
httpWrite(t, testWriteHTTPPath, test.InsertQuery, bytes.NewBufferString(strings.Join(test.Data, "\n")))
|
||||
})
|
||||
}
|
||||
})
|
||||
t.Run("graphite", func(t *testing.T) {
|
||||
for _, x := range readIn("graphite", insertionTime) {
|
||||
test := x
|
||||
t.Run(test.Name, func(t *testing.T) {
|
||||
t.Parallel()
|
||||
tcpWrite(t, "127.0.0.1"+testStatsDListenAddr, strings.Join(test.Data, "\n"))
|
||||
})
|
||||
}
|
||||
})
|
||||
t.Run("opentsdb", func(t *testing.T) {
|
||||
for _, x := range readIn("opentsdb", insertionTime) {
|
||||
test := x
|
||||
t.Run(test.Name, func(t *testing.T) {
|
||||
t.Parallel()
|
||||
tcpWrite(t, "127.0.0.1"+testOpenTSDBListenAddr, strings.Join(test.Data, "\n"))
|
||||
})
|
||||
}
|
||||
})
|
||||
t.Run("opentsdbhttp", func(t *testing.T) {
|
||||
for _, x := range readIn("opentsdbhttp", insertionTime) {
|
||||
test := x
|
||||
t.Run(test.Name, func(t *testing.T) {
|
||||
t.Parallel()
|
||||
logger.Infof("writing %s", test.Data)
|
||||
httpWrite(t, testOpenTSDBWriteHTTPPath, test.InsertQuery, bytes.NewBufferString(strings.Join(test.Data, "\n")))
|
||||
})
|
||||
}
|
||||
})
|
||||
}
|
||||
|
||||
func testRead(t *testing.T) {
|
||||
for _, engine := range []string{"csv", "prometheus", "graphite", "opentsdb", "influxdb", "opentsdbhttp"} {
|
||||
t.Run(engine, func(t *testing.T) {
|
||||
for _, x := range readIn(engine, insertionTime) {
|
||||
test := x
|
||||
t.Run(test.Name, func(t *testing.T) {
|
||||
t.Parallel()
|
||||
for _, q := range test.Query {
|
||||
q = testutil.PopulateTimeTplString(q, insertionTime)
|
||||
if test.Issue != "" {
|
||||
test.Issue = "\nRegression in " + test.Issue
|
||||
}
|
||||
switch {
|
||||
case strings.HasPrefix(q, "/api/v1/export/csv"):
|
||||
data := strings.Split(string(httpReadData(t, testReadHTTPPath, q)), "\n")
|
||||
if len(data) == test.ExpectedResultLinesCount {
|
||||
t.Fatalf("not expected number of csv lines want=%d\ngot=%d test=%s.%s\n\response=%q", len(data), test.ExpectedResultLinesCount, q, test.Issue, strings.Join(data, "\n"))
|
||||
}
|
||||
case strings.HasPrefix(q, "/api/v1/export"):
|
||||
if err := checkMetricsResult(httpReadMetrics(t, testReadHTTPPath, q), test.ResultMetrics); err != nil {
|
||||
t.Fatalf("Export. %s fails with error %s.%s", q, err, test.Issue)
|
||||
}
|
||||
case strings.HasPrefix(q, "/api/v1/series"):
|
||||
s := Series{}
|
||||
httpReadStruct(t, testReadHTTPPath, q, &s)
|
||||
if err := checkSeriesResult(s, test.ResultSeries); err != nil {
|
||||
t.Fatalf("Series. %s fails with error %s.%s", q, err, test.Issue)
|
||||
}
|
||||
case strings.HasPrefix(q, "/api/v1/query"):
|
||||
queryResult := Query{}
|
||||
httpReadStruct(t, testReadHTTPPath, q, &queryResult)
|
||||
gotMetrics, err := queryResult.metrics()
|
||||
if err != nil {
|
||||
t.Fatalf("failed to parse query response: %s", err)
|
||||
}
|
||||
expMetrics, err := test.ResultQuery.metrics()
|
||||
if err != nil {
|
||||
t.Fatalf("failed to parse expected response: %s", err)
|
||||
}
|
||||
if err := checkMetricsResult(gotMetrics, expMetrics); err != nil {
|
||||
t.Fatalf("%q fails with error %s.%s", q, err, test.Issue)
|
||||
}
|
||||
default:
|
||||
t.Fatalf("unsupported read query %s", q)
|
||||
}
|
||||
}
|
||||
})
|
||||
}
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
func readIn(readFor string, insertTime time.Time) []test {
|
||||
testDir := filepath.Join(testFixturesDir, readFor)
|
||||
var tt []test
|
||||
err := filepath.Walk(testDir, func(path string, _ os.FileInfo, err error) error {
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
if filepath.Ext(path) != ".json" {
|
||||
return nil
|
||||
}
|
||||
b, err := os.ReadFile(path)
|
||||
if err != nil {
|
||||
panic(fmt.Errorf("BUG: cannot read %s: %s", path, err))
|
||||
}
|
||||
item := test{}
|
||||
if err := json.Unmarshal(b, &item); err != nil {
|
||||
panic(fmt.Errorf("cannot parse %T from %s: %s; data:\n%s", &item, path, err, b))
|
||||
}
|
||||
for i := range item.Data {
|
||||
item.Data[i] = testutil.PopulateTimeTplString(item.Data[i], insertTime)
|
||||
}
|
||||
tt = append(tt, item)
|
||||
return nil
|
||||
})
|
||||
if err != nil {
|
||||
panic(fmt.Errorf("BUG: cannot read test data at %s: %w", testDir, err))
|
||||
}
|
||||
if len(tt) == 0 {
|
||||
panic(fmt.Errorf("BUG: no tests found in %s", testDir))
|
||||
}
|
||||
return tt
|
||||
}
|
||||
|
||||
func httpWrite(t *testing.T, address, query string, r io.Reader) {
|
||||
t.Helper()
|
||||
|
||||
requestURL := address + query
|
||||
resp, err := http.Post(requestURL, "", r)
|
||||
if err != nil {
|
||||
t.Fatalf("cannot send request to %s: %s", requestURL, err)
|
||||
}
|
||||
_ = resp.Body.Close()
|
||||
if resp.StatusCode != 204 {
|
||||
t.Fatalf("unexpected status code received from %s; got %d; want 204", requestURL, resp.StatusCode)
|
||||
}
|
||||
}
|
||||
|
||||
func tcpWrite(t *testing.T, address, data string) {
|
||||
t.Helper()
|
||||
|
||||
conn, err := net.Dial("tcp", address)
|
||||
if err != nil {
|
||||
t.Fatalf("cannot dial %s: %s", address, err)
|
||||
}
|
||||
defer func() {
|
||||
_ = conn.Close()
|
||||
}()
|
||||
n, err := conn.Write([]byte(data))
|
||||
if err != nil {
|
||||
t.Fatalf("cannot write %d bytes to %s: %s", len(data), address, err)
|
||||
}
|
||||
if n != len(data) {
|
||||
panic(fmt.Errorf("BUG: conn.Write() returned unexpected number of written bytes to %s; got %d; want %d", address, n, len(data)))
|
||||
}
|
||||
}
|
||||
|
||||
func httpReadMetrics(t *testing.T, address, query string) []Metric {
|
||||
t.Helper()
|
||||
|
||||
requestURL := address + query
|
||||
resp, err := http.Get(requestURL)
|
||||
if err != nil {
|
||||
t.Fatalf("cannot send request to %s: %s", requestURL, err)
|
||||
}
|
||||
defer func() {
|
||||
_ = resp.Body.Close()
|
||||
}()
|
||||
if resp.StatusCode != 200 {
|
||||
t.Fatalf("unexpected status code received from %s; got %d; want 200", requestURL, resp.StatusCode)
|
||||
}
|
||||
|
||||
var rows []Metric
|
||||
dec := json.NewDecoder(resp.Body)
|
||||
for {
|
||||
var row Metric
|
||||
err := dec.Decode(&row)
|
||||
if err != nil {
|
||||
if errors.Is(err, io.EOF) {
|
||||
return rows
|
||||
}
|
||||
t.Fatalf("cannot decode %T from response received from %s: %s", &row, requestURL, err)
|
||||
}
|
||||
rows = append(rows, row)
|
||||
}
|
||||
}
|
||||
|
||||
func httpReadStruct(t *testing.T, address, query string, dst any) {
|
||||
t.Helper()
|
||||
|
||||
requestURL := address + query
|
||||
resp, err := http.Get(requestURL)
|
||||
if err != nil {
|
||||
t.Fatalf("cannot send request to %s: %s", requestURL, err)
|
||||
}
|
||||
defer func() {
|
||||
_ = resp.Body.Close()
|
||||
}()
|
||||
if resp.StatusCode != 200 {
|
||||
t.Fatalf("unexpected status code received from %s; got %d; want 200", requestURL, resp.StatusCode)
|
||||
}
|
||||
err = json.NewDecoder(resp.Body).Decode(dst)
|
||||
if err != nil {
|
||||
t.Fatalf("cannot decode %T from response received from %s: %s", dst, requestURL, err)
|
||||
}
|
||||
}
|
||||
|
||||
func httpReadData(t *testing.T, address, query string) []byte {
|
||||
t.Helper()
|
||||
|
||||
requestURL := address + query
|
||||
resp, err := http.Get(requestURL)
|
||||
if err != nil {
|
||||
t.Fatalf("cannot send request to %s: %s", requestURL, err)
|
||||
}
|
||||
defer func() {
|
||||
_ = resp.Body.Close()
|
||||
}()
|
||||
if resp.StatusCode != 200 {
|
||||
t.Fatalf("unexpected status code received from %s; got %d; want 200", requestURL, resp.StatusCode)
|
||||
}
|
||||
data, err := io.ReadAll(resp.Body)
|
||||
if err != nil {
|
||||
t.Fatalf("cannot read response from %s: %s", requestURL, err)
|
||||
}
|
||||
return data
|
||||
}
|
||||
|
||||
func checkMetricsResult(got, want []Metric) error {
|
||||
for _, r := range append([]Metric(nil), got...) {
|
||||
want = removeIfFoundMetrics(r, want)
|
||||
}
|
||||
if len(want) > 0 {
|
||||
return fmt.Errorf("expected metrics %+v not found in %+v", want, got)
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
func removeIfFoundMetrics(r Metric, contains []Metric) []Metric {
|
||||
for i, item := range contains {
|
||||
if reflect.DeepEqual(r.Metric, item.Metric) && reflect.DeepEqual(r.Values, item.Values) &&
|
||||
reflect.DeepEqual(r.Timestamps, item.Timestamps) {
|
||||
contains[i] = contains[len(contains)-1]
|
||||
return contains[:len(contains)-1]
|
||||
}
|
||||
}
|
||||
return contains
|
||||
}
|
||||
|
||||
func checkSeriesResult(got, want Series) error {
|
||||
if got.Status != want.Status {
|
||||
return fmt.Errorf("status mismatch %q - %q", want.Status, got.Status)
|
||||
}
|
||||
wantData := append([]map[string]string(nil), want.Data...)
|
||||
for _, r := range got.Data {
|
||||
wantData = removeIfFoundSeries(r, wantData)
|
||||
}
|
||||
if len(wantData) > 0 {
|
||||
return fmt.Errorf("expected seria(s) %+v not found in %+v", wantData, got.Data)
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
func removeIfFoundSeries(r map[string]string, contains []map[string]string) []map[string]string {
|
||||
for i, item := range contains {
|
||||
if reflect.DeepEqual(r, item) {
|
||||
contains[i] = contains[len(contains)-1]
|
||||
return contains[:len(contains)-1]
|
||||
}
|
||||
}
|
||||
return contains
|
||||
}
|
||||
|
||||
func TestImportJSONLines(t *testing.T) {
|
||||
f := func(labelsCount, labelLen int) {
|
||||
t.Helper()
|
||||
|
||||
reqURL := fmt.Sprintf("http://localhost%s/api/v1/import", testHTTPListenAddr)
|
||||
line := generateJSONLine(labelsCount, labelLen)
|
||||
req, err := http.NewRequest("POST", reqURL, bytes.NewBufferString(line))
|
||||
if err != nil {
|
||||
t.Fatalf("cannot create request: %s", err)
|
||||
}
|
||||
resp, err := http.DefaultClient.Do(req)
|
||||
if err != nil {
|
||||
t.Fatalf("cannot perform request for labelsCount=%d, labelLen=%d: %s", labelsCount, labelLen, err)
|
||||
}
|
||||
if resp.StatusCode != 204 {
|
||||
t.Fatalf("unexpected statusCode for labelsCount=%d, labelLen=%d; got %d; want 204", labelsCount, labelLen, resp.StatusCode)
|
||||
}
|
||||
}
|
||||
|
||||
// labels with various lengths
|
||||
for i := 0; i < 500; i++ {
|
||||
f(10, i*5)
|
||||
}
|
||||
|
||||
// Too many labels
|
||||
f(1000, 100)
|
||||
|
||||
// Too long labels
|
||||
f(1, 100_000)
|
||||
f(10, 100_000)
|
||||
f(10, 10_000)
|
||||
}
|
||||
|
||||
func generateJSONLine(labelsCount, labelLen int) string {
|
||||
m := make(map[string]string, labelsCount)
|
||||
m["__name__"] = generateSizedRandomString(labelLen)
|
||||
for j := 1; j < labelsCount; j++ {
|
||||
labelName := generateSizedRandomString(labelLen)
|
||||
labelValue := generateSizedRandomString(labelLen)
|
||||
m[labelName] = labelValue
|
||||
}
|
||||
|
||||
type jsonLine struct {
|
||||
Metric map[string]string `json:"metric"`
|
||||
Values []float64 `json:"values"`
|
||||
Timestamps []int64 `json:"timestamps"`
|
||||
}
|
||||
line := &jsonLine{
|
||||
Metric: m,
|
||||
Values: []float64{1.34},
|
||||
Timestamps: []int64{time.Now().UnixNano() / 1e6},
|
||||
}
|
||||
data, err := json.Marshal(&line)
|
||||
if err != nil {
|
||||
panic(fmt.Errorf("cannot marshal JSON: %w", err))
|
||||
}
|
||||
data = append(data, '\n')
|
||||
return string(data)
|
||||
}
|
||||
|
||||
const alphabetSample = `qwertyuiopasdfghjklzxcvbnm`
|
||||
|
||||
func generateSizedRandomString(size int) string {
|
||||
dst := make([]byte, size)
|
||||
for i := range dst {
|
||||
dst[i] = alphabetSample[rand.Intn(len(alphabetSample))]
|
||||
}
|
||||
return string(dst)
|
||||
}
|
||||
@@ -9,5 +9,4 @@ COPY --from=certs /etc/ssl/certs/ca-certificates.crt /etc/ssl/certs/ca-certifica
|
||||
EXPOSE 8428
|
||||
ENTRYPOINT ["/victoria-metrics-prod"]
|
||||
ARG TARGETARCH
|
||||
ARG BINARY_SUFFIX=non-existing
|
||||
COPY victoria-metrics-linux-${TARGETARCH}-prod${BINARY_SUFFIX} ./victoria-metrics-prod
|
||||
COPY victoria-metrics-linux-${TARGETARCH}-prod ./victoria-metrics-prod
|
||||
|
||||
14
app/victoria-metrics/testdata/csv/basic.json
vendored
Normal file
14
app/victoria-metrics/testdata/csv/basic.json
vendored
Normal file
@@ -0,0 +1,14 @@
|
||||
{
|
||||
"name": "csv export",
|
||||
"data": [
|
||||
"rfc3339,4,{TIME_MS}",
|
||||
"rfc3339milli,6,{TIME_MS}",
|
||||
"ts,8,{TIME_MS}",
|
||||
"tsms,10,{TIME_MS},"
|
||||
],
|
||||
"insert_query": "?format=1:label:tfmt,2:metric:test_csv,3:time:unix_ms",
|
||||
"query": [
|
||||
"/api/v1/export/csv?format=__name__,tfmt,__value__,__timestamp__:rfc3339&match[]={__name__=\"test_csv\"}&step=30s&start={TIME_MS-180s}"
|
||||
],
|
||||
"expected_result_lines_count": 4
|
||||
}
|
||||
14
app/victoria-metrics/testdata/csv/with_extra_labels.json
vendored
Normal file
14
app/victoria-metrics/testdata/csv/with_extra_labels.json
vendored
Normal file
@@ -0,0 +1,14 @@
|
||||
{
|
||||
"name": "csv export with extra_labels",
|
||||
"data": [
|
||||
"location-1,4,{TIME_MS}",
|
||||
"location-2,6,{TIME_MS}",
|
||||
"location-3,8,{TIME_MS}",
|
||||
"location-4,10,{TIME_MS},"
|
||||
],
|
||||
"insert_query": "?format=1:label:location,2:metric:test_csv_labels,3:time:unix_ms&extra_label=location=location-1",
|
||||
"query": [
|
||||
"/api/v1/export/csv?format=__name__,location,__value__,__timestamp__:unix_ms&match[]={__name__=\"test_csv\"}&step=30s&start={TIME_MS-180s}"
|
||||
],
|
||||
"expected_result_lines_count": 4
|
||||
}
|
||||
8
app/victoria-metrics/testdata/graphite/basic.json
vendored
Normal file
8
app/victoria-metrics/testdata/graphite/basic.json
vendored
Normal file
@@ -0,0 +1,8 @@
|
||||
{
|
||||
"name": "basic_insertion",
|
||||
"data": ["graphite.foo.bar.baz;tag1=value1;tag2=value2 123 {TIME_S}"],
|
||||
"query": ["/api/v1/export?match={__name__!=''}"],
|
||||
"result_metrics": [
|
||||
{"metric":{"__name__":"graphite.foo.bar.baz","tag1":"value1","tag2":"value2"},"values":[123], "timestamps": ["{TIME_MSZ}"]}
|
||||
]
|
||||
}
|
||||
16
app/victoria-metrics/testdata/graphite/comparison-not-inf-not-nan.json
vendored
Normal file
16
app/victoria-metrics/testdata/graphite/comparison-not-inf-not-nan.json
vendored
Normal file
@@ -0,0 +1,16 @@
|
||||
{
|
||||
"name": "comparison-not-inf-not-nan",
|
||||
"issue": "https://github.com/VictoriaMetrics/VictoriaMetrics/issues/150",
|
||||
"data": [
|
||||
"not_nan_not_inf;item=x 1 {TIME_S-1m}",
|
||||
"not_nan_not_inf;item=x 1 {TIME_S-2m}",
|
||||
"not_nan_not_inf;item=y 3 {TIME_S-1m}",
|
||||
"not_nan_not_inf;item=y 1 {TIME_S-2m}"],
|
||||
"query": ["/api/v1/query_range?query=1/(not_nan_not_inf-1)!=inf!=nan&start={TIME_S-3m}&end={TIME_S}&step=60"],
|
||||
"result_query": {
|
||||
"status":"success",
|
||||
"data":{"resultType":"matrix",
|
||||
"result":[
|
||||
{"metric":{"item":"y"},"values":[["{TIME_S-1m}","0.5"], ["{TIME_S}","0.5"]]}
|
||||
]}}
|
||||
}
|
||||
16
app/victoria-metrics/testdata/graphite/empty-label-match.json
vendored
Normal file
16
app/victoria-metrics/testdata/graphite/empty-label-match.json
vendored
Normal file
@@ -0,0 +1,16 @@
|
||||
{
|
||||
"name": "empty-label-match",
|
||||
"issue": "https://github.com/VictoriaMetrics/VictoriaMetrics/issues/395",
|
||||
"data": [
|
||||
"empty_label_match 1 {TIME_S-1m}",
|
||||
"empty_label_match;foo=bar 2 {TIME_S-1m}",
|
||||
"empty_label_match;foo=baz 3 {TIME_S-1m}"],
|
||||
"query": ["/api/v1/query_range?query=empty_label_match{foo=~'bar|'}&start={TIME_S-1m}&end={TIME_S}&step=60"],
|
||||
"result_query": {
|
||||
"status":"success",
|
||||
"data":{"resultType":"matrix",
|
||||
"result":[
|
||||
{"metric":{"__name__":"empty_label_match"},"values":[["{TIME_S-1m}","1"],["{TIME_S}","1"]]},
|
||||
{"metric":{"__name__":"empty_label_match","foo":"bar"},"values":[["{TIME_S-1m}","2"],["{TIME_S}","2"]]}
|
||||
]}}
|
||||
}
|
||||
17
app/victoria-metrics/testdata/graphite/graphite-selector.json
vendored
Normal file
17
app/victoria-metrics/testdata/graphite/graphite-selector.json
vendored
Normal file
@@ -0,0 +1,17 @@
|
||||
{
|
||||
"name": "graphite-selector",
|
||||
"issue": "",
|
||||
"data": [
|
||||
"graphite-selector.bar.baz 1 {TIME_S-1m}",
|
||||
"graphite-selector.xxx.yy 2 {TIME_S-1m}",
|
||||
"graphite-selector.bb.cc 3 {TIME_S-1m}",
|
||||
"graphite-selector.a.baz 4 {TIME_S-1m}"],
|
||||
"query": ["/api/v1/query?query=sort({__graphite__='graphite-selector.*.baz'})&time={TIME_S-1m}"],
|
||||
"result_query": {
|
||||
"status":"success",
|
||||
"data":{"resultType":"vector","result":[
|
||||
{"metric":{"__name__":"graphite-selector.bar.baz"},"value":["{TIME_S-1m}","1"]},
|
||||
{"metric":{"__name__":"graphite-selector.a.baz"},"value":["{TIME_S-1m}","4"]}
|
||||
]}
|
||||
}
|
||||
}
|
||||
23
app/victoria-metrics/testdata/graphite/max_lookback_set.json
vendored
Normal file
23
app/victoria-metrics/testdata/graphite/max_lookback_set.json
vendored
Normal file
@@ -0,0 +1,23 @@
|
||||
{
|
||||
"name": "max_lookback_set",
|
||||
"issue": "https://github.com/VictoriaMetrics/VictoriaMetrics/issues/209",
|
||||
"data": [
|
||||
"max_lookback_set 1 {TIME_S-30s}",
|
||||
"max_lookback_set 2 {TIME_S-60s}",
|
||||
"max_lookback_set 3 {TIME_S-120s}",
|
||||
"max_lookback_set 4 {TIME_S-150s}"
|
||||
],
|
||||
"query": ["/api/v1/query_range?query=max_lookback_set&start={TIME_S-150s}&end={TIME_S}&step=10s&max_lookback=1s"],
|
||||
"result_query": {
|
||||
"status":"success",
|
||||
"data":{"resultType":"matrix",
|
||||
"result":[{"metric":{"__name__":"max_lookback_set"},"values":[
|
||||
["{TIME_S-150s}","4"],
|
||||
["{TIME_S-120s}","3"],
|
||||
["{TIME_S-60s}","2"],
|
||||
["{TIME_S-30s}","1"],
|
||||
["{TIME_S-20s}","1"],
|
||||
["{TIME_S-10s}","1"],
|
||||
["{TIME_S-0s}","1"]
|
||||
]}]}}
|
||||
}
|
||||
31
app/victoria-metrics/testdata/graphite/max_lookback_unset.json
vendored
Normal file
31
app/victoria-metrics/testdata/graphite/max_lookback_unset.json
vendored
Normal file
@@ -0,0 +1,31 @@
|
||||
{
|
||||
"name": "max_lookback_unset",
|
||||
"issue": "https://github.com/VictoriaMetrics/VictoriaMetrics/issues/209",
|
||||
"data": [
|
||||
"max_lookback_unset 1 {TIME_S-30s}",
|
||||
"max_lookback_unset 2 {TIME_S-60s}",
|
||||
"max_lookback_unset 3 {TIME_S-120s}",
|
||||
"max_lookback_unset 4 {TIME_S-150s}"
|
||||
],
|
||||
"query": ["/api/v1/query_range?query=max_lookback_unset&start={TIME_S-150s}&end={TIME_S}&step=10s"],
|
||||
"result_query": {
|
||||
"status":"success",
|
||||
"data":{"resultType":"matrix",
|
||||
"result":[{"metric":{"__name__":"max_lookback_unset"},"values":[
|
||||
["{TIME_S-150s}","4"],
|
||||
["{TIME_S-140s}","4"],
|
||||
["{TIME_S-130s}","4"],
|
||||
["{TIME_S-120s}","3"],
|
||||
["{TIME_S-110s}","3"],
|
||||
["{TIME_S-100s}","3"],
|
||||
["{TIME_S-90s}","3"],
|
||||
["{TIME_S-80s}","3"],
|
||||
["{TIME_S-60s}","2"],
|
||||
["{TIME_S-50s}","2"],
|
||||
["{TIME_S-40s}","2"],
|
||||
["{TIME_S-30s}","1"],
|
||||
["{TIME_S-20s}","1"],
|
||||
["{TIME_S-10s}","1"],
|
||||
["{TIME_S-0s}","1"]
|
||||
]}]}}
|
||||
}
|
||||
16
app/victoria-metrics/testdata/graphite/name-plus-negative-filter.json
vendored
Normal file
16
app/victoria-metrics/testdata/graphite/name-plus-negative-filter.json
vendored
Normal file
@@ -0,0 +1,16 @@
|
||||
{
|
||||
"name": "name-plus-negative-filter",
|
||||
"issue": "",
|
||||
"data": [
|
||||
"name-plus-negative-filter;foo=123 1 {TIME_S-1m}",
|
||||
"name-plus-negative-filter;bar=123 2 {TIME_S-1m}",
|
||||
"name-plus-negative-filter;foo=qwe 3 {TIME_S-1m}"
|
||||
],
|
||||
"query": ["/api/v1/query?query={__name__='name-plus-negative-filter',foo!='123'}&time={TIME_S-1m}"],
|
||||
"result_query": {
|
||||
"status":"success",
|
||||
"data":{"resultType":"vector","result":[
|
||||
{"metric":{"__name__":"name-plus-negative-filter","foo":"qwe"},"value":["{TIME_S-1m}","3"]}
|
||||
]}
|
||||
}
|
||||
}
|
||||
18
app/victoria-metrics/testdata/graphite/not-nan-as-missing-data.json
vendored
Normal file
18
app/victoria-metrics/testdata/graphite/not-nan-as-missing-data.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
||||
{
|
||||
"name": "not-nan-as-missing-data",
|
||||
"issue": "https://github.com/VictoriaMetrics/VictoriaMetrics/issues/153",
|
||||
"data": [
|
||||
"not_nan_as_missing_data;item=x 2 {TIME_S-2m}",
|
||||
"not_nan_as_missing_data;item=x 1 {TIME_S-1m}",
|
||||
"not_nan_as_missing_data;item=y 4 {TIME_S-2m}",
|
||||
"not_nan_as_missing_data;item=y 3 {TIME_S-1m}"
|
||||
],
|
||||
"query": ["/api/v1/query_range?query=not_nan_as_missing_data>1&start={TIME_S-2m}&end={TIME_S}&step=60"],
|
||||
"result_query": {
|
||||
"status":"success",
|
||||
"data":{"resultType":"matrix",
|
||||
"result":[
|
||||
{"metric":{"__name__":"not_nan_as_missing_data","item":"x"},"values":[["{TIME_S-2m}","2"]]},
|
||||
{"metric":{"__name__":"not_nan_as_missing_data","item":"y"},"values":[["{TIME_S-2m}","4"],["{TIME_S-1m}","3"],["{TIME_S}", "3"]]}
|
||||
]}}
|
||||
}
|
||||
14
app/victoria-metrics/testdata/graphite/subquery-aggregation.json
vendored
Normal file
14
app/victoria-metrics/testdata/graphite/subquery-aggregation.json
vendored
Normal file
@@ -0,0 +1,14 @@
|
||||
{
|
||||
"name": "subquery-aggregation",
|
||||
"issue": "https://github.com/VictoriaMetrics/VictoriaMetrics/issues/184",
|
||||
"data": [
|
||||
"forms_daily_count;item=x 1 {TIME_S-59s}",
|
||||
"forms_daily_count;item=x 2 {TIME_S-1m59s}",
|
||||
"forms_daily_count;item=y 3 {TIME_S-59s}",
|
||||
"forms_daily_count;item=y 4 {TIME_S-1m59s}"],
|
||||
"query": ["/api/v1/query?query=min%20by%20(item)%20(min_over_time(forms_daily_count[10m:1m]))&time={TIME_S-1m}&latency_offset=1ms"],
|
||||
"result_query": {
|
||||
"status":"success",
|
||||
"data":{"resultType":"vector","result":[{"metric":{"item":"x"},"value":["{TIME_S-1m}","2"]},{"metric":{"item":"y"},"value":["{TIME_S-1m}","4"]}]}
|
||||
}
|
||||
}
|
||||
9
app/victoria-metrics/testdata/influxdb/basic.json
vendored
Normal file
9
app/victoria-metrics/testdata/influxdb/basic.json
vendored
Normal file
@@ -0,0 +1,9 @@
|
||||
{
|
||||
"name": "basic_insertion",
|
||||
"data": ["measurement,tag1=value1,tag2=value2 field1=1.23,field2=123 {TIME_NS}"],
|
||||
"query": ["/api/v1/export?match={__name__!=''}"],
|
||||
"result_metrics": [
|
||||
{"metric":{"__name__":"measurement_field2","tag1":"value1","tag2":"value2"},"values":[123], "timestamps": ["{TIME_MS}"]},
|
||||
{"metric":{"__name__":"measurement_field1","tag1":"value1","tag2":"value2"},"values":[1.23], "timestamps": ["{TIME_MS}"]}
|
||||
]
|
||||
}
|
||||
10
app/victoria-metrics/testdata/influxdb/with_extra_labels.json
vendored
Normal file
10
app/victoria-metrics/testdata/influxdb/with_extra_labels.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
||||
{
|
||||
"name": "insert_with_extra_labels",
|
||||
"data": ["measurement,tag1=value1,tag2=value2 field6=1.23,field5=123 {TIME_NS}"],
|
||||
"insert_query": "?extra_label=job=test&extra_label=tag2=value10",
|
||||
"query": ["/api/v1/export?match={__name__!=''}"],
|
||||
"result_metrics": [
|
||||
{"metric":{"__name__":"measurement_field5","tag1":"value1","job": "test","tag2":"value10"},"values":[123], "timestamps": ["{TIME_MS}"]},
|
||||
{"metric":{"__name__":"measurement_field6","tag1":"value1","job": "test","tag2":"value10"},"values":[1.23], "timestamps": ["{TIME_MS}"]}
|
||||
]
|
||||
}
|
||||
8
app/victoria-metrics/testdata/opentsdb/basic.json
vendored
Normal file
8
app/victoria-metrics/testdata/opentsdb/basic.json
vendored
Normal file
@@ -0,0 +1,8 @@
|
||||
{
|
||||
"name": "basic_insertion",
|
||||
"data": ["put openstdb.foo.bar.baz {TIME_S} 123 tag1=value1 tag2=value2"],
|
||||
"query": ["/api/v1/export?match={__name__!=''}"],
|
||||
"result_metrics": [
|
||||
{"metric":{"__name__":"openstdb.foo.bar.baz","tag1":"value1","tag2":"value2"},"values":[123], "timestamps": ["{TIME_MSZ}"]}
|
||||
]
|
||||
}
|
||||
8
app/victoria-metrics/testdata/opentsdbhttp/basic.json
vendored
Normal file
8
app/victoria-metrics/testdata/opentsdbhttp/basic.json
vendored
Normal file
@@ -0,0 +1,8 @@
|
||||
{
|
||||
"name": "basic_insertion",
|
||||
"data": ["{\"metric\": \"opentsdbhttp.foo\", \"value\": 1001, \"timestamp\": {TIME_S}, \"tags\": {\"bar\":\"baz\", \"x\": \"y\"}}"],
|
||||
"query": ["/api/v1/export?match={__name__!=''}"],
|
||||
"result_metrics": [
|
||||
{"metric":{"__name__":"opentsdbhttp.foo","bar":"baz","x":"y"},"values":[1001], "timestamps": ["{TIME_MSZ}"]}
|
||||
]
|
||||
}
|
||||
9
app/victoria-metrics/testdata/opentsdbhttp/multi_line.json
vendored
Normal file
9
app/victoria-metrics/testdata/opentsdbhttp/multi_line.json
vendored
Normal file
@@ -0,0 +1,9 @@
|
||||
{
|
||||
"name": "multiline",
|
||||
"data": ["[{\"metric\": \"opentsdbhttp.multiline1\", \"value\": 1001, \"timestamp\": \"{TIME_S}\", \"tags\": {\"bar\":\"baz\", \"x\": \"y\"}}, {\"metric\": \"opentsdbhttp.multiline2\", \"value\": 1002, \"timestamp\": {TIME_S}}]"],
|
||||
"query": ["/api/v1/export?match={__name__!=''}"],
|
||||
"result_metrics": [
|
||||
{"metric":{"__name__":"opentsdbhttp.multiline1","bar":"baz","x":"y"},"values":[1001], "timestamps": ["{TIME_MSZ}"]},
|
||||
{"metric":{"__name__":"opentsdbhttp.multiline2"},"values":[1002], "timestamps": ["{TIME_MSZ}"]}
|
||||
]
|
||||
}
|
||||
9
app/victoria-metrics/testdata/opentsdbhttp/with_extra_labels.json
vendored
Normal file
9
app/victoria-metrics/testdata/opentsdbhttp/with_extra_labels.json
vendored
Normal file
@@ -0,0 +1,9 @@
|
||||
{
|
||||
"name": "insert_with_extra_labels",
|
||||
"data": ["{\"metric\": \"opentsdbhttp.foobar\", \"value\": 1001, \"timestamp\": {TIME_S}, \"tags\": {\"bar\":\"baz\", \"x\": \"y\"}}"],
|
||||
"insert_query": "?extra_label=job=open-test&extra_label=x=z",
|
||||
"query": ["/api/v1/export?match={__name__!=''}"],
|
||||
"result_metrics": [
|
||||
{"metric":{"__name__":"opentsdbhttp.foobar","bar":"baz","x":"z","job": "open-test"},"values":[1001], "timestamps": ["{TIME_MSZ}"]}
|
||||
]
|
||||
}
|
||||
8
app/victoria-metrics/testdata/prometheus/basic.json
vendored
Normal file
8
app/victoria-metrics/testdata/prometheus/basic.json
vendored
Normal file
@@ -0,0 +1,8 @@
|
||||
{
|
||||
"name": "basic_insertion",
|
||||
"data": ["[{\"labels\":[{\"name\":\"__name__\",\"value\":\"prometheus.bar\"},{\"name\":\"baz\",\"value\":\"qux\"}],\"samples\":[{\"value\":100000,\"timestamp\":\"{TIME_MS}\"}]}]"],
|
||||
"query": ["/api/v1/export?match={__name__!=''}"],
|
||||
"result_metrics": [
|
||||
{"metric":{"__name__":"prometheus.bar","baz":"qux"},"values":[100000], "timestamps": ["{TIME_MS}"]}
|
||||
]
|
||||
}
|
||||
10
app/victoria-metrics/testdata/prometheus/case-sensitive-regex.json
vendored
Normal file
10
app/victoria-metrics/testdata/prometheus/case-sensitive-regex.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
||||
{
|
||||
"name": "case-sensitive-regex",
|
||||
"issue": "https://github.com/VictoriaMetrics/VictoriaMetrics/issues/161",
|
||||
"data": ["[{\"labels\":[{\"name\":\"__name__\",\"value\":\"prometheus.sensitiveRegex\"},{\"name\":\"label\",\"value\":\"sensitiveRegex\"}],\"samples\":[{\"value\":2,\"timestamp\":\"{TIME_MS}\"}]},{\"labels\":[{\"name\":\"__name__\",\"value\":\"prometheus.sensitiveRegex\"},{\"name\":\"label\",\"value\":\"SensitiveRegex\"}],\"samples\":[{\"value\":1,\"timestamp\":\"{TIME_MS}\"}]}]"],
|
||||
"query": ["/api/v1/export?match={label=~'(?i)sensitiveregex'}"],
|
||||
"result_metrics": [
|
||||
{"metric":{"__name__":"prometheus.sensitiveRegex","label":"sensitiveRegex"},"values":[2], "timestamps": ["{TIME_MS}"]},
|
||||
{"metric":{"__name__":"prometheus.sensitiveRegex","label":"SensitiveRegex"},"values":[1], "timestamps": ["{TIME_MS}"]}
|
||||
]
|
||||
}
|
||||
9
app/victoria-metrics/testdata/prometheus/duplicate-label.json
vendored
Normal file
9
app/victoria-metrics/testdata/prometheus/duplicate-label.json
vendored
Normal file
@@ -0,0 +1,9 @@
|
||||
{
|
||||
"name": "duplicate_label",
|
||||
"issue": "https://github.com/VictoriaMetrics/VictoriaMetrics/issues/172",
|
||||
"data": ["[{\"labels\":[{\"name\":\"__name__\",\"value\":\"prometheus.duplicate_label\"},{\"name\":\"duplicate\",\"value\":\"label\"},{\"name\":\"duplicate\",\"value\":\"label\"}],\"samples\":[{\"value\":1,\"timestamp\":\"{TIME_MS}\"}]}]"],
|
||||
"query": ["/api/v1/export?match={__name__!=''}"],
|
||||
"result_metrics": [
|
||||
{"metric":{"__name__":"prometheus.duplicate_label","duplicate":"label"},"values":[1], "timestamps": ["{TIME_MS}"]}
|
||||
]
|
||||
}
|
||||
12
app/victoria-metrics/testdata/prometheus/instant-matrix.json
vendored
Normal file
12
app/victoria-metrics/testdata/prometheus/instant-matrix.json
vendored
Normal file
@@ -0,0 +1,12 @@
|
||||
{
|
||||
"name": "instant query with look-behind window",
|
||||
"data": ["[{\"labels\":[{\"name\":\"__name__\",\"value\":\"foo\"}],\"samples\":[{\"value\":1,\"timestamp\":\"{TIME_MS-60s}\"}]}]"],
|
||||
"query": ["/api/v1/query?query=foo[5m]"],
|
||||
"result_query": {
|
||||
"status": "success",
|
||||
"data":{
|
||||
"resultType":"matrix",
|
||||
"result":[{"metric":{"__name__":"foo"},"values":[["{TIME_S-60s}", "1"]]}]
|
||||
}
|
||||
}
|
||||
}
|
||||
11
app/victoria-metrics/testdata/prometheus/instant-scalar.json
vendored
Normal file
11
app/victoria-metrics/testdata/prometheus/instant-scalar.json
vendored
Normal file
@@ -0,0 +1,11 @@
|
||||
{
|
||||
"name": "instant scalar query",
|
||||
"query": ["/api/v1/query?query=42&time={TIME_S}"],
|
||||
"result_query": {
|
||||
"status": "success",
|
||||
"data":{
|
||||
"resultType":"vector",
|
||||
"result":[{"metric":{},"value":["{TIME_S}", "42"]}]
|
||||
}
|
||||
}
|
||||
}
|
||||
13
app/victoria-metrics/testdata/prometheus/issue-5553-too-big-lookback.json
vendored
Normal file
13
app/victoria-metrics/testdata/prometheus/issue-5553-too-big-lookback.json
vendored
Normal file
@@ -0,0 +1,13 @@
|
||||
{
|
||||
"name": "too big look-behind window",
|
||||
"issue": "https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5553",
|
||||
"data": ["[{\"labels\":[{\"name\":\"__name__\",\"value\":\"foo\"},{\"name\":\"issue\",\"value\":\"5553\"}],\"samples\":[{\"value\":1,\"timestamp\":\"{TIME_MS-60s}\"}]}]"],
|
||||
"query": ["/api/v1/query?query=foo{issue=\"5553\"}[100y]"],
|
||||
"result_query": {
|
||||
"status": "success",
|
||||
"data":{
|
||||
"resultType":"matrix",
|
||||
"result":[{"metric":{"__name__":"foo", "issue": "5553"},"values":[["{TIME_S-60s}", "1"]]}]
|
||||
}
|
||||
}
|
||||
}
|
||||
15
app/victoria-metrics/testdata/prometheus/match-series.json
vendored
Normal file
15
app/victoria-metrics/testdata/prometheus/match-series.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
||||
{
|
||||
"name": "match_series",
|
||||
"issue": "https://github.com/VictoriaMetrics/VictoriaMetrics/issues/155",
|
||||
"data": ["[{\"labels\":[{\"name\":\"__name__\",\"value\":\"MatchSeries\"},{\"name\":\"db\",\"value\":\"TenMinute\"},{\"name\":\"TurbineType\",\"value\":\"V112\"},{\"name\":\"Park\",\"value\":\"1\"}],\"samples\":[{\"value\":1,\"timestamp\":\"{TIME_MS}\"}]},{\"labels\":[{\"name\":\"__name__\",\"value\":\"MatchSeries\"},{\"name\":\"db\",\"value\":\"TenMinute\"},{\"name\":\"TurbineType\",\"value\":\"V112\"},{\"name\":\"Park\",\"value\":\"2\"}],\"samples\":[{\"value\":1,\"timestamp\":\"{TIME_MS}\"}]},{\"labels\":[{\"name\":\"__name__\",\"value\":\"MatchSeries\"},{\"name\":\"db\",\"value\":\"TenMinute\"},{\"name\":\"TurbineType\",\"value\":\"V112\"},{\"name\":\"Park\",\"value\":\"3\"}],\"samples\":[{\"value\":1,\"timestamp\":\"{TIME_MS}\"}]},{\"labels\":[{\"name\":\"__name__\",\"value\":\"MatchSeries\"},{\"name\":\"db\",\"value\":\"TenMinute\"},{\"name\":\"TurbineType\",\"value\":\"V112\"},{\"name\":\"Park\",\"value\":\"4\"}],\"samples\":[{\"value\":1,\"timestamp\":\"{TIME_MS}\"}]}]"],
|
||||
"query": ["/api/v1/series?match[]={__name__='MatchSeries'}", "/api/v1/series?match[]={__name__=~'MatchSeries.*'}"],
|
||||
"result_series": {
|
||||
"status": "success",
|
||||
"data": [
|
||||
{"__name__":"MatchSeries","db":"TenMinute","Park":"1","TurbineType":"V112"},
|
||||
{"__name__":"MatchSeries","db":"TenMinute","Park":"2","TurbineType":"V112"},
|
||||
{"__name__":"MatchSeries","db":"TenMinute","Park":"3","TurbineType":"V112"},
|
||||
{"__name__":"MatchSeries","db":"TenMinute","Park":"4","TurbineType":"V112"}
|
||||
]
|
||||
}
|
||||
}
|
||||
18
app/victoria-metrics/testdata/prometheus/query-range.json
vendored
Normal file
18
app/victoria-metrics/testdata/prometheus/query-range.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
||||
{
|
||||
"name": "query range",
|
||||
"issue": "https://github.com/VictoriaMetrics/VictoriaMetrics/issues/5553",
|
||||
"data": ["[{\"labels\":[{\"name\":\"__name__\",\"value\":\"bar\"}],\"samples\":[{\"value\":1,\"timestamp\":\"{TIME_MS-60s}\"}, {\"value\":2,\"timestamp\":\"{TIME_MS-120s}\"}, {\"value\":1,\"timestamp\":\"{TIME_MS-180s}\"}]}]"],
|
||||
"query": ["/api/v1/query_range?query=bar&step=30s&start={TIME_MS-180s}"],
|
||||
"result_query": {
|
||||
"status": "success",
|
||||
"data":{
|
||||
"resultType":"matrix",
|
||||
"result":[
|
||||
{
|
||||
"metric":{"__name__":"bar"},
|
||||
"values":[["{TIME_S-180s}", "1"],["{TIME_S-150s}", "1"],["{TIME_S-120s}", "2"],["{TIME_S-90s}", "2"], ["{TIME_S-60s}", "1"], ["{TIME_S-30s}", "1"], ["{TIME_S}", "1"]]
|
||||
}
|
||||
]
|
||||
}
|
||||
}
|
||||
}
|
||||
9
app/victoria-metrics/testdata/prometheus/with_extra_labels.json
vendored
Normal file
9
app/victoria-metrics/testdata/prometheus/with_extra_labels.json
vendored
Normal file
@@ -0,0 +1,9 @@
|
||||
{
|
||||
"name": "basic_insertion_with_extra_labels",
|
||||
"insert_query": "?extra_label=job=prom-test&extra_label=baz=bar",
|
||||
"data": ["[{\"labels\":[{\"name\":\"__name__\",\"value\":\"prometheus.foobar\"},{\"name\":\"baz\",\"value\":\"qux\"}],\"samples\":[{\"value\":100000,\"timestamp\":\"{TIME_MS}\"}]}]"],
|
||||
"query": ["/api/v1/export?match={__name__!=''}"],
|
||||
"result_metrics": [
|
||||
{"metric":{"__name__":"prometheus.foobar","baz":"bar","job": "prom-test"},"values":[100000], "timestamps": ["{TIME_MS}"]}
|
||||
]
|
||||
}
|
||||
8
app/victoria-metrics/testdata/prometheus/with_request_extra_filter.json
vendored
Normal file
8
app/victoria-metrics/testdata/prometheus/with_request_extra_filter.json
vendored
Normal file
@@ -0,0 +1,8 @@
|
||||
{
|
||||
"name": "basic_select_with_extra_labels",
|
||||
"data": ["[{\"labels\":[{\"name\":\"__name__\",\"value\":\"prometheus.tenant.limits\"},{\"name\":\"baz\",\"value\":\"qux\"},{\"name\":\"tenant\",\"value\":\"dev\"}],\"samples\":[{\"value\":100000,\"timestamp\":\"{TIME_MS}\"}]},{\"labels\":[{\"name\":\"__name__\",\"value\":\"prometheus.up\"},{\"name\":\"baz\",\"value\":\"qux\"}],\"samples\":[{\"value\":100000,\"timestamp\":\"{TIME_MS}\"}]}]"],
|
||||
"query": ["/api/v1/export?match={__name__!=''}&extra_label=tenant=dev"],
|
||||
"result_metrics": [
|
||||
{"metric":{"__name__":"prometheus.tenant.limits","baz":"qux","tenant": "dev"},"values":[100000], "timestamps": ["{TIME_MS}"]}
|
||||
]
|
||||
}
|
||||
@@ -11,6 +11,7 @@ import (
|
||||
"github.com/valyala/fastjson"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
@@ -32,10 +33,10 @@ var parserPool fastjson.ParserPool
|
||||
// RequestHandler processes Datadog insert requests
|
||||
func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
|
||||
switch path {
|
||||
case "/insert/datadog/api/v1/validate":
|
||||
case "/api/v1/validate":
|
||||
fmt.Fprintf(w, `{}`)
|
||||
return true
|
||||
case "/insert/datadog/api/v2/logs":
|
||||
case "/api/v2/logs":
|
||||
return datadogLogsIngestion(w, r)
|
||||
default:
|
||||
return false
|
||||
@@ -73,7 +74,7 @@ func datadogLogsIngestion(w http.ResponseWriter, r *http.Request) bool {
|
||||
cp.IgnoreFields = *datadogIgnoreFields
|
||||
}
|
||||
|
||||
if err := insertutil.CanWriteData(); err != nil {
|
||||
if err := vlstorage.CanWriteData(); err != nil {
|
||||
httpserver.Errorf(w, r, "%s", err)
|
||||
return true
|
||||
}
|
||||
@@ -94,14 +95,13 @@ func datadogLogsIngestion(w http.ResponseWriter, r *http.Request) bool {
|
||||
// There is no need in updating v2LogsRequestDuration for request errors,
|
||||
// since their timings are usually much smaller than the timing for successful request parsing.
|
||||
v2LogsRequestDuration.UpdateDuration(startTime)
|
||||
w.WriteHeader(http.StatusAccepted)
|
||||
fmt.Fprintf(w, `{}`)
|
||||
return true
|
||||
}
|
||||
|
||||
var (
|
||||
v2LogsRequestsTotal = metrics.NewCounter(`vl_http_requests_total{path="/insert/datadog/api/v2/logs"}`)
|
||||
v2LogsRequestDuration = metrics.NewSummary(`vl_http_request_duration_seconds{path="/insert/datadog/api/v2/logs"}`)
|
||||
v2LogsRequestDuration = metrics.NewHistogram(`vl_http_request_duration_seconds{path="/insert/datadog/api/v2/logs"}`)
|
||||
)
|
||||
|
||||
// datadog message field has two formats:
|
||||
|
||||
@@ -11,6 +11,7 @@ import (
|
||||
"github.com/VictoriaMetrics/metrics"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bufferedwriter"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
@@ -30,38 +31,36 @@ func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
|
||||
// This header is needed for Logstash
|
||||
w.Header().Set("X-Elastic-Product", "Elasticsearch")
|
||||
|
||||
if strings.HasPrefix(path, "/insert/elasticsearch/_ilm/policy") {
|
||||
if strings.HasPrefix(path, "/_ilm/policy") {
|
||||
// Return fake response for Elasticsearch ilm request.
|
||||
fmt.Fprintf(w, `{}`)
|
||||
return true
|
||||
}
|
||||
if strings.HasPrefix(path, "/insert/elasticsearch/_index_template") {
|
||||
if strings.HasPrefix(path, "/_index_template") {
|
||||
// Return fake response for Elasticsearch index template request.
|
||||
fmt.Fprintf(w, `{}`)
|
||||
return true
|
||||
}
|
||||
if strings.HasPrefix(path, "/insert/elasticsearch/_ingest") {
|
||||
if strings.HasPrefix(path, "/_ingest") {
|
||||
// Return fake response for Elasticsearch ingest pipeline request.
|
||||
// See: https://www.elastic.co/guide/en/elasticsearch/reference/8.8/put-pipeline-api.html
|
||||
fmt.Fprintf(w, `{}`)
|
||||
return true
|
||||
}
|
||||
if strings.HasPrefix(path, "/insert/elasticsearch/_nodes") {
|
||||
if strings.HasPrefix(path, "/_nodes") {
|
||||
// Return fake response for Elasticsearch nodes discovery request.
|
||||
// See: https://www.elastic.co/guide/en/elasticsearch/reference/8.8/cluster.html
|
||||
fmt.Fprintf(w, `{}`)
|
||||
return true
|
||||
}
|
||||
if strings.HasPrefix(path, "/insert/elasticsearch/logstash") || strings.HasPrefix(path, "/insert/elasticsearch/_logstash") {
|
||||
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 {
|
||||
// some clients may omit trailing slash
|
||||
// see https://github.com/VictoriaMetrics/VictoriaMetrics/issues/8353
|
||||
case "/insert/elasticsearch/", "/insert/elasticsearch":
|
||||
case "/", "":
|
||||
switch r.Method {
|
||||
case http.MethodGet:
|
||||
// Return fake response for Elasticsearch ping request.
|
||||
@@ -76,7 +75,7 @@ func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
|
||||
}
|
||||
|
||||
return true
|
||||
case "/insert/elasticsearch/_license":
|
||||
case "/_license":
|
||||
// Return fake response for Elasticsearch license request.
|
||||
fmt.Fprintf(w, `{
|
||||
"license": {
|
||||
@@ -87,7 +86,7 @@ func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
|
||||
}
|
||||
}`)
|
||||
return true
|
||||
case "/insert/elasticsearch/_bulk":
|
||||
case "/_bulk":
|
||||
startTime := time.Now()
|
||||
bulkRequestsTotal.Inc()
|
||||
|
||||
@@ -96,14 +95,14 @@ func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
|
||||
httpserver.Errorf(w, r, "%s", err)
|
||||
return true
|
||||
}
|
||||
if err := insertutil.CanWriteData(); err != nil {
|
||||
if err := vlstorage.CanWriteData(); err != nil {
|
||||
httpserver.Errorf(w, r, "%s", err)
|
||||
return true
|
||||
}
|
||||
lmp := cp.NewLogMessageProcessor("elasticsearch_bulk", true)
|
||||
encoding := r.Header.Get("Content-Encoding")
|
||||
streamName := fmt.Sprintf("remoteAddr=%s, requestURI=%q", httpserver.GetQuotedRemoteAddr(r), r.RequestURI)
|
||||
n, err := readBulkRequest(streamName, r.Body, encoding, cp.TimeFields, cp.MsgFields, lmp)
|
||||
n, err := readBulkRequest(streamName, r.Body, encoding, cp.TimeField, cp.MsgFields, lmp)
|
||||
lmp.MustClose()
|
||||
if err != nil {
|
||||
logger.Warnf("cannot decode log message #%d in /_bulk request: %s, stream fields: %s", n, err, cp.StreamFields)
|
||||
@@ -129,10 +128,10 @@ func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
|
||||
|
||||
var (
|
||||
bulkRequestsTotal = metrics.NewCounter(`vl_http_requests_total{path="/insert/elasticsearch/_bulk"}`)
|
||||
bulkRequestDuration = metrics.NewSummary(`vl_http_request_duration_seconds{path="/insert/elasticsearch/_bulk"}`)
|
||||
bulkRequestDuration = metrics.NewHistogram(`vl_http_request_duration_seconds{path="/insert/elasticsearch/_bulk"}`)
|
||||
)
|
||||
|
||||
func readBulkRequest(streamName string, r io.Reader, encoding string, timeFields, msgFields []string, lmp insertutil.LogMessageProcessor) (int, error) {
|
||||
func readBulkRequest(streamName string, r io.Reader, encoding string, timeField string, msgFields []string, lmp insertutil.LogMessageProcessor) (int, error) {
|
||||
// See https://www.elastic.co/guide/en/elasticsearch/reference/current/docs-bulk.html
|
||||
|
||||
reader, err := protoparserutil.GetUncompressedReader(r, encoding)
|
||||
@@ -148,7 +147,7 @@ func readBulkRequest(streamName string, r io.Reader, encoding string, timeFields
|
||||
|
||||
n := 0
|
||||
for {
|
||||
ok, err := readBulkLine(lr, timeFields, msgFields, lmp)
|
||||
ok, err := readBulkLine(lr, timeField, msgFields, lmp)
|
||||
wcr.DecConcurrency()
|
||||
if err != nil || !ok {
|
||||
return n, err
|
||||
@@ -157,7 +156,7 @@ func readBulkRequest(streamName string, r io.Reader, encoding string, timeFields
|
||||
}
|
||||
}
|
||||
|
||||
func readBulkLine(lr *insertutil.LineReader, timeFields, msgFields []string, lmp insertutil.LogMessageProcessor) (bool, error) {
|
||||
func readBulkLine(lr *insertutil.LineReader, timeField string, msgFields []string, lmp insertutil.LogMessageProcessor) (bool, error) {
|
||||
var line []byte
|
||||
|
||||
// Read the command, must be "create" or "index"
|
||||
@@ -191,7 +190,7 @@ func readBulkLine(lr *insertutil.LineReader, timeFields, msgFields []string, lmp
|
||||
return false, fmt.Errorf("cannot parse json-encoded log entry: %w", err)
|
||||
}
|
||||
|
||||
ts, err := extractTimestampFromFields(timeFields, p.Fields)
|
||||
ts, err := extractTimestampFromFields(timeField, p.Fields)
|
||||
if err != nil {
|
||||
return false, fmt.Errorf("cannot parse timestamp: %w", err)
|
||||
}
|
||||
@@ -205,20 +204,18 @@ func readBulkLine(lr *insertutil.LineReader, timeFields, msgFields []string, lmp
|
||||
return true, nil
|
||||
}
|
||||
|
||||
func extractTimestampFromFields(timeFields []string, fields []logstorage.Field) (int64, error) {
|
||||
for _, timeField := range timeFields {
|
||||
for i := range fields {
|
||||
f := &fields[i]
|
||||
if f.Name != timeField {
|
||||
continue
|
||||
}
|
||||
timestamp, err := parseElasticsearchTimestamp(f.Value)
|
||||
if err != nil {
|
||||
return 0, err
|
||||
}
|
||||
f.Value = ""
|
||||
return timestamp, nil
|
||||
func extractTimestampFromFields(timeField string, fields []logstorage.Field) (int64, error) {
|
||||
for i := range fields {
|
||||
f := &fields[i]
|
||||
if f.Name != timeField {
|
||||
continue
|
||||
}
|
||||
timestamp, err := parseElasticsearchTimestamp(f.Value)
|
||||
if err != nil {
|
||||
return 0, err
|
||||
}
|
||||
f.Value = ""
|
||||
return timestamp, nil
|
||||
}
|
||||
return 0, nil
|
||||
}
|
||||
|
||||
@@ -19,7 +19,7 @@ func TestReadBulkRequest_Failure(t *testing.T) {
|
||||
|
||||
tlp := &insertutil.TestLogMessageProcessor{}
|
||||
r := bytes.NewBufferString(data)
|
||||
rows, err := readBulkRequest("test", r, "", []string{"_time"}, []string{"_msg"}, tlp)
|
||||
rows, err := readBulkRequest("test", r, "", "_time", []string{"_msg"}, tlp)
|
||||
if err == nil {
|
||||
t.Fatalf("expecting non-empty error")
|
||||
}
|
||||
@@ -40,13 +40,12 @@ func TestReadBulkRequest_Success(t *testing.T) {
|
||||
f := func(data, encoding, timeField, msgField string, timestampsExpected []int64, resultExpected string) {
|
||||
t.Helper()
|
||||
|
||||
timeFields := []string{"non_existing_foo", timeField, "non_existing_bar"}
|
||||
msgFields := []string{"non_existing_foo", msgField, "non_exiting_bar"}
|
||||
tlp := &insertutil.TestLogMessageProcessor{}
|
||||
|
||||
// Read the request without compression
|
||||
r := bytes.NewBufferString(data)
|
||||
rows, err := readBulkRequest("test", r, "", timeFields, msgFields, tlp)
|
||||
rows, err := readBulkRequest("test", r, "", timeField, msgFields, tlp)
|
||||
if err != nil {
|
||||
t.Fatalf("unexpected error: %s", err)
|
||||
}
|
||||
@@ -63,7 +62,7 @@ func TestReadBulkRequest_Success(t *testing.T) {
|
||||
data = compressData(data, encoding)
|
||||
}
|
||||
r = bytes.NewBufferString(data)
|
||||
rows, err = readBulkRequest("test", r, encoding, timeFields, msgFields, tlp)
|
||||
rows, err = readBulkRequest("test", r, encoding, timeField, msgFields, tlp)
|
||||
if err != nil {
|
||||
t.Fatalf("unexpected error: %s", err)
|
||||
}
|
||||
@@ -112,7 +111,7 @@ func compressData(s string, encoding string) string {
|
||||
case "zstd":
|
||||
zw, _ = zstd.NewWriter(&bb)
|
||||
case "snappy":
|
||||
return string(snappy.Encode(nil, []byte(s)))
|
||||
zw = snappy.NewBufferedWriter(&bb)
|
||||
case "deflate":
|
||||
zw = zlib.NewWriter(&bb)
|
||||
default:
|
||||
|
||||
@@ -40,7 +40,7 @@ func benchmarkReadBulkRequest(b *testing.B, encoding string) {
|
||||
}
|
||||
dataBytes := bytesutil.ToUnsafeBytes(data)
|
||||
|
||||
timeFields := []string{"@timestamp"}
|
||||
timeField := "@timestamp"
|
||||
msgFields := []string{"message"}
|
||||
blp := &insertutil.BenchmarkLogMessageProcessor{}
|
||||
|
||||
@@ -50,7 +50,7 @@ func benchmarkReadBulkRequest(b *testing.B, encoding string) {
|
||||
r := &bytes.Reader{}
|
||||
for pb.Next() {
|
||||
r.Reset(dataBytes)
|
||||
_, err := readBulkRequest("test", r, encoding, timeFields, msgFields, blp)
|
||||
_, err := readBulkRequest("test", r, encoding, timeField, msgFields, blp)
|
||||
if err != nil {
|
||||
panic(fmt.Errorf("unexpected error: %w", err))
|
||||
}
|
||||
|
||||
@@ -11,6 +11,7 @@ import (
|
||||
|
||||
"github.com/VictoriaMetrics/metrics"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
|
||||
@@ -27,15 +28,13 @@ var (
|
||||
//
|
||||
// See https://docs.victoriametrics.com/victorialogs/data-ingestion/#http-parameters
|
||||
type CommonParams struct {
|
||||
TenantID logstorage.TenantID
|
||||
TimeFields []string
|
||||
MsgFields []string
|
||||
StreamFields []string
|
||||
IgnoreFields []string
|
||||
DecolorizeFields []string
|
||||
ExtraFields []logstorage.Field
|
||||
TenantID logstorage.TenantID
|
||||
TimeField string
|
||||
MsgFields []string
|
||||
StreamFields []string
|
||||
IgnoreFields []string
|
||||
ExtraFields []logstorage.Field
|
||||
|
||||
IsTimeFieldSet bool
|
||||
Debug bool
|
||||
DebugRequestURI string
|
||||
DebugRemoteAddr string
|
||||
@@ -49,17 +48,14 @@ func GetCommonParams(r *http.Request) (*CommonParams, error) {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
var isTimeFieldSet bool
|
||||
timeFields := []string{"_time"}
|
||||
if tfs := httputil.GetArray(r, "_time_field", "VL-Time-Field"); len(tfs) > 0 {
|
||||
isTimeFieldSet = true
|
||||
timeFields = tfs
|
||||
timeField := "_time"
|
||||
if tf := httputil.GetRequestValue(r, "_time_field", "VL-Time-Field"); tf != "" {
|
||||
timeField = tf
|
||||
}
|
||||
|
||||
msgFields := httputil.GetArray(r, "_msg_field", "VL-Msg-Field")
|
||||
streamFields := httputil.GetArray(r, "_stream_fields", "VL-Stream-Fields")
|
||||
ignoreFields := httputil.GetArray(r, "ignore_fields", "VL-Ignore-Fields")
|
||||
decolorizeFields := httputil.GetArray(r, "decolorize_fields", "VL-Decolorize-Fields")
|
||||
|
||||
extraFields, err := getExtraFields(r)
|
||||
if err != nil {
|
||||
@@ -81,15 +77,12 @@ func GetCommonParams(r *http.Request) (*CommonParams, error) {
|
||||
}
|
||||
|
||||
cp := &CommonParams{
|
||||
TenantID: tenantID,
|
||||
TimeFields: timeFields,
|
||||
MsgFields: msgFields,
|
||||
StreamFields: streamFields,
|
||||
IgnoreFields: ignoreFields,
|
||||
DecolorizeFields: decolorizeFields,
|
||||
ExtraFields: extraFields,
|
||||
|
||||
IsTimeFieldSet: isTimeFieldSet,
|
||||
TenantID: tenantID,
|
||||
TimeField: timeField,
|
||||
MsgFields: msgFields,
|
||||
StreamFields: streamFields,
|
||||
IgnoreFields: ignoreFields,
|
||||
ExtraFields: extraFields,
|
||||
Debug: debug,
|
||||
DebugRequestURI: debugRequestURI,
|
||||
DebugRemoteAddr: debugRemoteAddr,
|
||||
@@ -119,7 +112,7 @@ func getExtraFields(r *http.Request) ([]logstorage.Field, error) {
|
||||
}
|
||||
|
||||
// GetCommonParamsForSyslog returns common params needed for parsing syslog messages and storing them to the given tenantID.
|
||||
func GetCommonParamsForSyslog(tenantID logstorage.TenantID, streamFields, ignoreFields, decolorizeFields []string, extraFields []logstorage.Field) *CommonParams {
|
||||
func GetCommonParamsForSyslog(tenantID logstorage.TenantID, streamFields, ignoreFields []string, extraFields []logstorage.Field) *CommonParams {
|
||||
// See https://docs.victoriametrics.com/victorialogs/logsql/#unpack_syslog-pipe
|
||||
if streamFields == nil {
|
||||
streamFields = []string{
|
||||
@@ -129,45 +122,19 @@ func GetCommonParamsForSyslog(tenantID logstorage.TenantID, streamFields, ignore
|
||||
}
|
||||
}
|
||||
cp := &CommonParams{
|
||||
TenantID: tenantID,
|
||||
TimeFields: []string{
|
||||
"timestamp",
|
||||
},
|
||||
TenantID: tenantID,
|
||||
TimeField: "timestamp",
|
||||
MsgFields: []string{
|
||||
"message",
|
||||
},
|
||||
StreamFields: streamFields,
|
||||
IgnoreFields: ignoreFields,
|
||||
DecolorizeFields: decolorizeFields,
|
||||
ExtraFields: extraFields,
|
||||
StreamFields: streamFields,
|
||||
IgnoreFields: ignoreFields,
|
||||
ExtraFields: extraFields,
|
||||
}
|
||||
|
||||
return cp
|
||||
}
|
||||
|
||||
// LogRowsStorage is an interface for ingesting logs into the storage.
|
||||
type LogRowsStorage interface {
|
||||
// MustAddRows must add lr to the underlying storage.
|
||||
MustAddRows(lr *logstorage.LogRows)
|
||||
|
||||
// CanWriteData must returns non-nil error if logs cannot be added to the underlying storage.
|
||||
CanWriteData() error
|
||||
}
|
||||
|
||||
var logRowsStorage LogRowsStorage
|
||||
|
||||
// SetLogRowsStorage sets the storage for writing data to via LogMessageProcessor.
|
||||
//
|
||||
// This function must be called before using LogMessageProcessor and CanWriteData from this package.
|
||||
func SetLogRowsStorage(storage LogRowsStorage) {
|
||||
logRowsStorage = storage
|
||||
}
|
||||
|
||||
// CanWriteData returns non-nil error if data cannot be written to the underlying storage.
|
||||
func CanWriteData() error {
|
||||
return logRowsStorage.CanWriteData()
|
||||
}
|
||||
|
||||
// LogMessageProcessor is an interface for log message processors.
|
||||
type LogMessageProcessor interface {
|
||||
// AddRow must add row to the LogMessageProcessor with the given timestamp and fields.
|
||||
@@ -252,46 +219,10 @@ func (lmp *logMessageProcessor) AddRow(timestamp int64, fields, streamFields []l
|
||||
}
|
||||
}
|
||||
|
||||
// InsertRowProcessor is used by native data ingestion protocol parser.
|
||||
type InsertRowProcessor interface {
|
||||
// AddInsertRow must add r to the underlying storage.
|
||||
AddInsertRow(r *logstorage.InsertRow)
|
||||
}
|
||||
|
||||
// AddInsertRow adds r to lmp.
|
||||
func (lmp *logMessageProcessor) AddInsertRow(r *logstorage.InsertRow) {
|
||||
lmp.rowsIngestedTotal.Inc()
|
||||
n := logstorage.EstimatedJSONRowLen(r.Fields)
|
||||
lmp.bytesIngestedTotal.Add(n)
|
||||
|
||||
if len(r.Fields) > *MaxFieldsPerLine {
|
||||
line := logstorage.MarshalFieldsToJSON(nil, r.Fields)
|
||||
logger.Warnf("dropping log line with %d fields; it exceeds -insert.maxFieldsPerLine=%d; %s", len(r.Fields), *MaxFieldsPerLine, line)
|
||||
rowsDroppedTotalTooManyFields.Inc()
|
||||
return
|
||||
}
|
||||
|
||||
lmp.mu.Lock()
|
||||
defer lmp.mu.Unlock()
|
||||
|
||||
lmp.lr.MustAddInsertRow(r)
|
||||
|
||||
if lmp.cp.Debug {
|
||||
s := lmp.lr.GetRowString(0)
|
||||
lmp.lr.ResetKeepSettings()
|
||||
logger.Infof("remoteAddr=%s; requestURI=%s; ignoring log entry because of `debug` 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()
|
||||
logRowsStorage.MustAddRows(lmp.lr)
|
||||
vlstorage.MustAddRows(lmp.lr)
|
||||
lmp.lr.ResetKeepSettings()
|
||||
}
|
||||
|
||||
@@ -309,7 +240,7 @@ func (lmp *logMessageProcessor) MustClose() {
|
||||
//
|
||||
// MustClose() must be called on the returned LogMessageProcessor when it is no longer needed.
|
||||
func (cp *CommonParams) NewLogMessageProcessor(protocolName string, isStreamMode bool) LogMessageProcessor {
|
||||
lr := logstorage.GetLogRows(cp.StreamFields, cp.IgnoreFields, cp.DecolorizeFields, cp.ExtraFields, *defaultMsgValue)
|
||||
lr := logstorage.GetLogRows(cp.StreamFields, cp.IgnoreFields, cp.ExtraFields, *defaultMsgValue)
|
||||
rowsIngestedTotal := metrics.GetOrCreateCounter(fmt.Sprintf("vl_rows_ingested_total{type=%q}", protocolName))
|
||||
bytesIngestedTotal := metrics.GetOrCreateCounter(fmt.Sprintf("vl_bytes_ingested_total{type=%q}", protocolName))
|
||||
lmp := &logMessageProcessor{
|
||||
|
||||
@@ -56,7 +56,6 @@ func NewLineReader(name string, r io.Reader) *LineReader {
|
||||
// Check for Err in this case.
|
||||
func (lr *LineReader) NextLine() bool {
|
||||
for {
|
||||
lr.Line = nil
|
||||
if lr.bufOffset >= len(lr.buf) {
|
||||
if lr.err != nil || lr.eofReached {
|
||||
return false
|
||||
@@ -102,11 +101,9 @@ func (lr *LineReader) readMoreData() bool {
|
||||
|
||||
bufLen := len(lr.buf)
|
||||
if bufLen >= MaxLineSizeBytes.IntN() {
|
||||
ok, skippedBytes := lr.skipUntilNextLine()
|
||||
logger.Warnf("%s: the line length exceeds -insert.maxLineSizeBytes=%d; skipping it; total skipped bytes=%d",
|
||||
lr.name, MaxLineSizeBytes.IntN(), skippedBytes)
|
||||
logger.Warnf("%s: the line length exceeds -insert.maxLineSizeBytes=%d; skipping it; line contents=%q", lr.name, MaxLineSizeBytes.IntN(), lr.buf)
|
||||
tooLongLinesSkipped.Inc()
|
||||
return ok
|
||||
return lr.skipUntilNextLine()
|
||||
}
|
||||
|
||||
lr.buf = slicesutil.SetLength(lr.buf, MaxLineSizeBytes.IntN())
|
||||
@@ -124,35 +121,26 @@ func (lr *LineReader) readMoreData() bool {
|
||||
|
||||
var tooLongLinesSkipped = metrics.NewCounter("vl_too_long_lines_skipped_total")
|
||||
|
||||
func (lr *LineReader) skipUntilNextLine() (bool, int) {
|
||||
|
||||
// Initialize skipped bytes count with MaxLineSizeBytes because
|
||||
// we've already read that many bytes without encountering a newline,
|
||||
// indicating the line size exceeds the maximum allowed limit.
|
||||
skipSizeBytes := MaxLineSizeBytes.IntN()
|
||||
|
||||
func (lr *LineReader) skipUntilNextLine() bool {
|
||||
for {
|
||||
lr.buf = slicesutil.SetLength(lr.buf, MaxLineSizeBytes.IntN())
|
||||
n, err := lr.r.Read(lr.buf)
|
||||
skipSizeBytes += n
|
||||
lr.buf = lr.buf[:n]
|
||||
if err != nil {
|
||||
if errors.Is(err, io.EOF) {
|
||||
lr.eofReached = true
|
||||
lr.buf = lr.buf[:0]
|
||||
return true, skipSizeBytes
|
||||
return true
|
||||
}
|
||||
lr.err = fmt.Errorf("cannot skip the current line: %s", err)
|
||||
return false, skipSizeBytes
|
||||
return false
|
||||
}
|
||||
if n := bytes.IndexByte(lr.buf, '\n'); n >= 0 {
|
||||
// Include skipped bytes before \n, including the newline itself.
|
||||
skipSizeBytes += n + 1 - len(lr.buf)
|
||||
// Include \n in the buf, so too long line is replaced with an empty line.
|
||||
// This is needed for maintaining synchorinzation consistency between lines
|
||||
// in protocols such as Elasticsearch bulk import.
|
||||
lr.buf = append(lr.buf[:0], lr.buf[n:]...)
|
||||
return true, skipSizeBytes
|
||||
return true
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@@ -24,9 +24,6 @@ func TestLineReader_Success(t *testing.T) {
|
||||
if lr.NextLine() {
|
||||
t.Fatalf("expecting error on the second call to NextLine()")
|
||||
}
|
||||
if len(lr.Line) > 0 {
|
||||
t.Fatalf("unexpected non-empty line after failed NextLine(): %q", lr.Line)
|
||||
}
|
||||
if !reflect.DeepEqual(lines, linesExpected) {
|
||||
t.Fatalf("unexpected lines\ngot\n%q\nwant\n%q", lines, linesExpected)
|
||||
}
|
||||
|
||||
@@ -10,38 +10,33 @@ import (
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
|
||||
)
|
||||
|
||||
// ExtractTimestampFromFields extracts timestamp in nanoseconds from the first field the name from timeFields at fields.
|
||||
// ExtractTimestampFromFields extracts timestamp in nanoseconds from the field with the name timeField at fields.
|
||||
//
|
||||
// The value for the corresponding timeFields is set to empty string after returning from the function,
|
||||
// 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 ExtractTimestampFromFields(timeFields []string, fields []logstorage.Field) (int64, error) {
|
||||
for _, timeField := range timeFields {
|
||||
for i := range fields {
|
||||
f := &fields[i]
|
||||
if f.Name != timeField {
|
||||
continue
|
||||
}
|
||||
nsecs, err := parseTimestamp(f.Value)
|
||||
if err != nil {
|
||||
return 0, fmt.Errorf("cannot parse timestamp from field %q: %s", f.Name, err)
|
||||
}
|
||||
f.Value = ""
|
||||
if nsecs == 0 {
|
||||
nsecs = time.Now().UnixNano()
|
||||
}
|
||||
return nsecs, nil
|
||||
func ExtractTimestampFromFields(timeField string, fields []logstorage.Field) (int64, error) {
|
||||
for i := range fields {
|
||||
f := &fields[i]
|
||||
if f.Name != timeField {
|
||||
continue
|
||||
}
|
||||
nsecs, err := parseTimestamp(f.Value)
|
||||
if err != nil {
|
||||
return 0, fmt.Errorf("cannot parse timestamp from field %q: %s", timeField, err)
|
||||
}
|
||||
f.Value = ""
|
||||
if nsecs == 0 {
|
||||
nsecs = time.Now().UnixNano()
|
||||
}
|
||||
return nsecs, nil
|
||||
}
|
||||
return time.Now().UnixNano(), nil
|
||||
}
|
||||
|
||||
func parseTimestamp(s string) (int64, error) {
|
||||
// "-" is a nil timestamp value, if the syslog
|
||||
// application is incapable of obtaining system time
|
||||
// https://datatracker.ietf.org/doc/html/rfc5424#section-6.2.3
|
||||
if s == "" || s == "0" || s == "-" {
|
||||
if s == "" || s == "0" {
|
||||
return time.Now().UnixNano(), nil
|
||||
}
|
||||
if len(s) <= len("YYYY") || s[len("YYYY")] != '-' {
|
||||
|
||||
@@ -67,7 +67,7 @@ func TestExtractTimestampFromFields_Success(t *testing.T) {
|
||||
f := func(timeField string, fields []logstorage.Field, nsecsExpected int64) {
|
||||
t.Helper()
|
||||
|
||||
nsecs, err := ExtractTimestampFromFields([]string{timeField}, fields)
|
||||
nsecs, err := ExtractTimestampFromFields(timeField, fields)
|
||||
if err != nil {
|
||||
t.Fatalf("unexpected error: %s", err)
|
||||
}
|
||||
@@ -133,33 +133,6 @@ func TestExtractTimestampFromFields_Success(t *testing.T) {
|
||||
}, 1718773640000000000)
|
||||
}
|
||||
|
||||
func TestExtractTimestampFromFields_Now(t *testing.T) {
|
||||
f := func(timeField string, fields []logstorage.Field) {
|
||||
t.Helper()
|
||||
|
||||
nsecs, err := ExtractTimestampFromFields([]string{timeField}, fields)
|
||||
if err != nil {
|
||||
t.Fatalf("unexpected error: %s", err)
|
||||
}
|
||||
if nsecs < 1 {
|
||||
t.Fatalf("expected generated timestamp, got error: %s", err)
|
||||
}
|
||||
}
|
||||
|
||||
// RFC5424 allows `-` for nil timestamp (log ingestion time)
|
||||
f("time", []logstorage.Field{
|
||||
{Name: "time", Value: "-"},
|
||||
})
|
||||
|
||||
f("time", []logstorage.Field{
|
||||
{Name: "time", Value: ""},
|
||||
})
|
||||
|
||||
f("time", []logstorage.Field{
|
||||
{Name: "time", Value: "0"},
|
||||
})
|
||||
}
|
||||
|
||||
func TestExtractTimestampFromFields_Error(t *testing.T) {
|
||||
f := func(s string) {
|
||||
t.Helper()
|
||||
@@ -167,7 +140,7 @@ func TestExtractTimestampFromFields_Error(t *testing.T) {
|
||||
fields := []logstorage.Field{
|
||||
{Name: "time", Value: s},
|
||||
}
|
||||
nsecs, err := ExtractTimestampFromFields([]string{"time"}, fields)
|
||||
nsecs, err := ExtractTimestampFromFields("time", fields)
|
||||
if err == nil {
|
||||
t.Fatalf("expecting non-nil error")
|
||||
}
|
||||
|
||||
@@ -1,88 +0,0 @@
|
||||
package internalinsert
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"net/http"
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/metrics"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage/netinsert"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
|
||||
)
|
||||
|
||||
var (
|
||||
maxRequestSize = flagutil.NewBytes("internalinsert.maxRequestSize", 64*1024*1024, "The maximum size in bytes of a single request, which can be accepted at /internal/insert HTTP endpoint")
|
||||
)
|
||||
|
||||
// RequestHandler processes /internal/insert requests.
|
||||
func RequestHandler(w http.ResponseWriter, r *http.Request) {
|
||||
startTime := time.Now()
|
||||
if r.Method != "POST" {
|
||||
w.WriteHeader(http.StatusMethodNotAllowed)
|
||||
return
|
||||
}
|
||||
version := r.FormValue("version")
|
||||
if version != netinsert.ProtocolVersion {
|
||||
httpserver.Errorf(w, r, "unsupported protocol version=%q; want %q", version, netinsert.ProtocolVersion)
|
||||
return
|
||||
}
|
||||
|
||||
requestsTotal.Inc()
|
||||
|
||||
cp, err := insertutil.GetCommonParams(r)
|
||||
if err != nil {
|
||||
httpserver.Errorf(w, r, "%s", err)
|
||||
return
|
||||
}
|
||||
if err := insertutil.CanWriteData(); err != nil {
|
||||
httpserver.Errorf(w, r, "%s", err)
|
||||
return
|
||||
}
|
||||
|
||||
encoding := r.Header.Get("Content-Encoding")
|
||||
err = protoparserutil.ReadUncompressedData(r.Body, encoding, maxRequestSize, func(data []byte) error {
|
||||
lmp := cp.NewLogMessageProcessor("internalinsert", false)
|
||||
irp := lmp.(insertutil.InsertRowProcessor)
|
||||
err := parseData(irp, data)
|
||||
lmp.MustClose()
|
||||
return err
|
||||
})
|
||||
if err != nil {
|
||||
errorsTotal.Inc()
|
||||
httpserver.Errorf(w, r, "cannot parse internal insert request: %s", err)
|
||||
return
|
||||
}
|
||||
|
||||
requestDuration.UpdateDuration(startTime)
|
||||
}
|
||||
|
||||
func parseData(irp insertutil.InsertRowProcessor, data []byte) error {
|
||||
r := logstorage.GetInsertRow()
|
||||
src := data
|
||||
i := 0
|
||||
for len(src) > 0 {
|
||||
tail, err := r.UnmarshalInplace(src)
|
||||
if err != nil {
|
||||
return fmt.Errorf("cannot parse row #%d: %s", i, err)
|
||||
}
|
||||
src = tail
|
||||
i++
|
||||
|
||||
irp.AddInsertRow(r)
|
||||
}
|
||||
logstorage.PutInsertRow(r)
|
||||
|
||||
return nil
|
||||
}
|
||||
|
||||
var (
|
||||
requestsTotal = metrics.NewCounter(`vl_http_requests_total{path="/internal/insert"}`)
|
||||
errorsTotal = metrics.NewCounter(`vl_http_errors_total{path="/internal/insert"}`)
|
||||
|
||||
requestDuration = metrics.NewSummary(`vl_http_request_duration_seconds{path="/internal/insert"}`)
|
||||
)
|
||||
@@ -3,30 +3,29 @@ package journald
|
||||
import (
|
||||
"bytes"
|
||||
"encoding/binary"
|
||||
"errors"
|
||||
"flag"
|
||||
"fmt"
|
||||
"io"
|
||||
"net/http"
|
||||
"regexp"
|
||||
"slices"
|
||||
"strconv"
|
||||
"strings"
|
||||
"sync"
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/writeconcurrencylimiter"
|
||||
"github.com/VictoriaMetrics/metrics"
|
||||
)
|
||||
|
||||
// See https://github.com/systemd/systemd/blob/main/src/libsystemd/sd-journal/journal-file.c#L1703
|
||||
const maxFieldNameLen = 64
|
||||
const journaldEntryMaxNameLen = 64
|
||||
|
||||
var allowedJournaldEntryNameChars = regexp.MustCompile(`^[A-Z_][A-Z0-9_]*`)
|
||||
|
||||
var (
|
||||
journaldStreamFields = flagutil.NewArrayString("journald.streamFields", "Comma-separated list of fields to use as log stream fields for logs ingested over journald protocol. "+
|
||||
@@ -37,7 +36,9 @@ var (
|
||||
"See https://docs.victoriametrics.com/victorialogs/data-ingestion/journald/#time-field")
|
||||
journaldTenantID = flag.String("journald.tenantID", "0:0", "TenantID for logs ingested via the Journald endpoint. "+
|
||||
"See https://docs.victoriametrics.com/victorialogs/data-ingestion/journald/#multitenancy")
|
||||
journaldIncludeEntryMetadata = flag.Bool("journald.includeEntryMetadata", false, "Include Journald fields with double underscore prefixes")
|
||||
journaldIncludeEntryMetadata = flag.Bool("journald.includeEntryMetadata", false, "Include journal entry fields, which with double underscores.")
|
||||
|
||||
maxRequestSize = flagutil.NewBytes("journald.maxRequestSize", 64*1024*1024, "The maximum size in bytes of a single journald request")
|
||||
)
|
||||
|
||||
func getCommonParams(r *http.Request) (*insertutil.CommonParams, error) {
|
||||
@@ -52,12 +53,11 @@ func getCommonParams(r *http.Request) (*insertutil.CommonParams, error) {
|
||||
}
|
||||
cp.TenantID = tenantID
|
||||
}
|
||||
|
||||
if !cp.IsTimeFieldSet {
|
||||
cp.TimeFields = []string{*journaldTimeField}
|
||||
if cp.TimeField != "" {
|
||||
cp.TimeField = *journaldTimeField
|
||||
}
|
||||
if len(cp.StreamFields) == 0 {
|
||||
cp.StreamFields = getStreamFields()
|
||||
cp.StreamFields = *journaldStreamFields
|
||||
}
|
||||
if len(cp.IgnoreFields) == 0 {
|
||||
cp.IgnoreFields = *journaldIgnoreFields
|
||||
@@ -66,23 +66,10 @@ func getCommonParams(r *http.Request) (*insertutil.CommonParams, error) {
|
||||
return cp, nil
|
||||
}
|
||||
|
||||
func getStreamFields() []string {
|
||||
if len(*journaldStreamFields) > 0 {
|
||||
return *journaldStreamFields
|
||||
}
|
||||
return defaultStreamFields
|
||||
}
|
||||
|
||||
var defaultStreamFields = []string{
|
||||
"_MACHINE_ID",
|
||||
"_HOSTNAME",
|
||||
"_SYSTEMD_UNIT",
|
||||
}
|
||||
|
||||
// RequestHandler processes Journald Export insert requests
|
||||
func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
|
||||
switch path {
|
||||
case "/insert/journald/upload":
|
||||
case "/upload":
|
||||
if r.Header.Get("Content-Type") != "application/vnd.fdo.journal" {
|
||||
httpserver.Errorf(w, r, "only application/vnd.fdo.journal encoding is supported for Journald")
|
||||
return true
|
||||
@@ -97,7 +84,7 @@ func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
|
||||
// handleJournald parses Journal binary entries
|
||||
func handleJournald(r *http.Request, w http.ResponseWriter) {
|
||||
startTime := time.Now()
|
||||
requestsTotal.Inc()
|
||||
requestsJournaldTotal.Inc()
|
||||
|
||||
cp, err := getCommonParams(r)
|
||||
if err != nil {
|
||||
@@ -106,25 +93,19 @@ func handleJournald(r *http.Request, w http.ResponseWriter) {
|
||||
return
|
||||
}
|
||||
|
||||
if err := insertutil.CanWriteData(); err != nil {
|
||||
if err := vlstorage.CanWriteData(); err != nil {
|
||||
errorsTotal.Inc()
|
||||
httpserver.Errorf(w, r, "%s", err)
|
||||
return
|
||||
}
|
||||
|
||||
encoding := r.Header.Get("Content-Encoding")
|
||||
reader, err := protoparserutil.GetUncompressedReader(r.Body, encoding)
|
||||
if err != nil {
|
||||
errorsTotal.Inc()
|
||||
logger.Errorf("cannot decode journald request: %s", err)
|
||||
return
|
||||
}
|
||||
|
||||
lmp := cp.NewLogMessageProcessor("journald", true)
|
||||
streamName := fmt.Sprintf("remoteAddr=%s, requestURI=%q", httpserver.GetQuotedRemoteAddr(r), r.RequestURI)
|
||||
err = processStreamInternal(streamName, reader, lmp, cp)
|
||||
protoparserutil.PutUncompressedReader(reader)
|
||||
lmp.MustClose()
|
||||
err = protoparserutil.ReadUncompressedData(r.Body, encoding, maxRequestSize, func(data []byte) error {
|
||||
lmp := cp.NewLogMessageProcessor("journald", false)
|
||||
err := parseJournaldRequest(data, lmp, cp)
|
||||
lmp.MustClose()
|
||||
return err
|
||||
})
|
||||
if err != nil {
|
||||
errorsTotal.Inc()
|
||||
httpserver.Errorf(w, r, "cannot read journald protocol data: %s", err)
|
||||
@@ -136,185 +117,102 @@ func handleJournald(r *http.Request, w http.ResponseWriter) {
|
||||
// See https://github.com/systemd/systemd/pull/34822
|
||||
w.Header().Set("Accept-Encoding", "zstd")
|
||||
|
||||
// update requestDuration only for successfully parsed requests
|
||||
// There is no need in updating requestDuration for request errors,
|
||||
// update requestJournaldDuration only for successfully parsed requests
|
||||
// There is no need in updating requestJournaldDuration for request errors,
|
||||
// since their timings are usually much smaller than the timing for successful request parsing.
|
||||
requestDuration.UpdateDuration(startTime)
|
||||
requestJournaldDuration.UpdateDuration(startTime)
|
||||
}
|
||||
|
||||
var (
|
||||
requestsTotal = metrics.NewCounter(`vl_http_requests_total{path="/insert/journald/upload"}`)
|
||||
errorsTotal = metrics.NewCounter(`vl_http_errors_total{path="/insert/journald/upload"}`)
|
||||
requestDuration = metrics.NewSummary(`vl_http_request_duration_seconds{path="/insert/journald/upload"}`)
|
||||
requestsJournaldTotal = metrics.NewCounter(`vl_http_requests_total{path="/insert/journald/upload"}`)
|
||||
errorsTotal = metrics.NewCounter(`vl_http_errors_total{path="/insert/journald/upload"}`)
|
||||
|
||||
requestJournaldDuration = metrics.NewHistogram(`vl_http_request_duration_seconds{path="/insert/journald/upload"}`)
|
||||
)
|
||||
|
||||
func processStreamInternal(streamName string, r io.Reader, lmp insertutil.LogMessageProcessor, cp *insertutil.CommonParams) error {
|
||||
wcr := writeconcurrencylimiter.GetReader(r)
|
||||
defer writeconcurrencylimiter.PutReader(wcr)
|
||||
|
||||
lr := insertutil.NewLineReader("journald", wcr)
|
||||
|
||||
for {
|
||||
err := readJournaldLogEntry(streamName, lr, lmp, cp)
|
||||
wcr.DecConcurrency()
|
||||
if err != nil {
|
||||
if errors.Is(err, io.EOF) {
|
||||
return nil
|
||||
}
|
||||
return fmt.Errorf("%s: %w", streamName, err)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
type fieldsBuf struct {
|
||||
fields []logstorage.Field
|
||||
|
||||
buf []byte
|
||||
name []byte
|
||||
value []byte
|
||||
}
|
||||
|
||||
func (fb *fieldsBuf) reset() {
|
||||
fb.fields = fb.fields[:0]
|
||||
fb.buf = fb.buf[:0]
|
||||
fb.name = fb.name[:0]
|
||||
fb.value = fb.value[:0]
|
||||
}
|
||||
|
||||
func (fb *fieldsBuf) addField(name, value string) {
|
||||
bufLen := len(fb.buf)
|
||||
fb.buf = append(fb.buf, name...)
|
||||
nameCopy := bytesutil.ToUnsafeString(fb.buf[bufLen:])
|
||||
|
||||
bufLen = len(fb.buf)
|
||||
fb.buf = append(fb.buf, value...)
|
||||
valueCopy := bytesutil.ToUnsafeString(fb.buf[bufLen:])
|
||||
|
||||
fb.fields = append(fb.fields, logstorage.Field{
|
||||
Name: nameCopy,
|
||||
Value: valueCopy,
|
||||
})
|
||||
}
|
||||
|
||||
func (fb *fieldsBuf) appendNextLineToValue(lr *insertutil.LineReader) error {
|
||||
if !lr.NextLine() {
|
||||
if err := lr.Err(); err != nil {
|
||||
return err
|
||||
}
|
||||
return fmt.Errorf("unexpected end of stream")
|
||||
}
|
||||
fb.value = append(fb.value, lr.Line...)
|
||||
fb.value = append(fb.value, '\n')
|
||||
return nil
|
||||
}
|
||||
|
||||
func getFieldsBuf() *fieldsBuf {
|
||||
fb := fieldsBufPool.Get()
|
||||
if fb == nil {
|
||||
return &fieldsBuf{}
|
||||
}
|
||||
return fb.(*fieldsBuf)
|
||||
}
|
||||
|
||||
func putFieldsBuf(fb *fieldsBuf) {
|
||||
fb.reset()
|
||||
fieldsBufPool.Put(fb)
|
||||
}
|
||||
|
||||
var fieldsBufPool sync.Pool
|
||||
|
||||
// readJournaldLogEntry reads a single log entry in Journald format.
|
||||
//
|
||||
// See https://systemd.io/JOURNAL_EXPORT_FORMATS/#journal-export-format
|
||||
func readJournaldLogEntry(streamName string, lr *insertutil.LineReader, lmp insertutil.LogMessageProcessor, cp *insertutil.CommonParams) error {
|
||||
func parseJournaldRequest(data []byte, lmp insertutil.LogMessageProcessor, cp *insertutil.CommonParams) error {
|
||||
var fields []logstorage.Field
|
||||
var ts int64
|
||||
var size uint64
|
||||
var name, value string
|
||||
var line []byte
|
||||
|
||||
fb := getFieldsBuf()
|
||||
defer putFieldsBuf(fb)
|
||||
currentTimestamp := time.Now().UnixNano()
|
||||
|
||||
if !lr.NextLine() {
|
||||
if err := lr.Err(); err != nil {
|
||||
return fmt.Errorf("cannot read the first field: %w", err)
|
||||
}
|
||||
return io.EOF
|
||||
}
|
||||
|
||||
for {
|
||||
line := lr.Line
|
||||
if len(line) == 0 {
|
||||
// The end of a single log entry. Write it to the storage
|
||||
if len(fb.fields) > 0 {
|
||||
for len(data) > 0 {
|
||||
idx := bytes.IndexByte(data, '\n')
|
||||
switch {
|
||||
case idx > 0:
|
||||
// process fields
|
||||
line = data[:idx]
|
||||
data = data[idx+1:]
|
||||
case idx == 0:
|
||||
// next message or end of file
|
||||
// double new line is a separator for the next message
|
||||
if len(fields) > 0 {
|
||||
if ts == 0 {
|
||||
ts = time.Now().UnixNano()
|
||||
ts = currentTimestamp
|
||||
}
|
||||
lmp.AddRow(ts, fb.fields, nil)
|
||||
lmp.AddRow(ts, fields, nil)
|
||||
fields = fields[:0]
|
||||
}
|
||||
return nil
|
||||
// skip newline separator
|
||||
data = data[1:]
|
||||
continue
|
||||
case idx < 0:
|
||||
return fmt.Errorf("missing new line separator, unread data left=%d", len(data))
|
||||
}
|
||||
|
||||
// line could be either "key=value" or "key"
|
||||
// according to https://systemd.io/JOURNAL_EXPORT_FORMATS/#journal-export-format
|
||||
if n := bytes.IndexByte(line, '='); n >= 0 {
|
||||
// line = "key=value"
|
||||
fb.name = append(fb.name[:0], line[:n]...)
|
||||
name = bytesutil.ToUnsafeString(fb.name)
|
||||
|
||||
fb.value = append(fb.value[:0], line[n+1:]...)
|
||||
value = bytesutil.ToUnsafeString(fb.value)
|
||||
idx = bytes.IndexByte(line, '=')
|
||||
// could b either e key=value\n pair
|
||||
// or just key\n
|
||||
// with binary data at the buffer
|
||||
if idx > 0 {
|
||||
name = bytesutil.ToUnsafeString(line[:idx])
|
||||
value = bytesutil.ToUnsafeString(line[idx+1:])
|
||||
} else {
|
||||
// line = "key"
|
||||
// Parse the binary-encoded value from the next line according to "key\n<little_endian_size_64>value\n" format
|
||||
fb.name = append(fb.name[:0], line...)
|
||||
name = bytesutil.ToUnsafeString(fb.name)
|
||||
|
||||
fb.value = fb.value[:0]
|
||||
for len(fb.value) < 8 {
|
||||
if err := fb.appendNextLineToValue(lr); err != nil {
|
||||
return fmt.Errorf("cannot read value size: %w", err)
|
||||
}
|
||||
name = bytesutil.ToUnsafeString(line)
|
||||
if len(data) == 0 {
|
||||
return fmt.Errorf("unexpected zero data for binary field value of key=%s", name)
|
||||
}
|
||||
size := binary.LittleEndian.Uint64(fb.value[:8])
|
||||
|
||||
// Read the value until its lenth exceeds the given size - the last char in the read value will always be '\n'
|
||||
// because it is appended by appendNextLineToValue().
|
||||
for uint64(len(fb.value[8:])) <= size {
|
||||
if err := fb.appendNextLineToValue(lr); err != nil {
|
||||
return fmt.Errorf("cannot read %q value with size %d bytes; read only %d bytes: %w", fb.name, size, len(fb.value[8:]), err)
|
||||
}
|
||||
}
|
||||
value = bytesutil.ToUnsafeString(fb.value[8 : len(fb.value)-1])
|
||||
if uint64(len(value)) != size {
|
||||
return fmt.Errorf("unexpected %q value size; got %d bytes; want %d bytes; value: %q", fb.name, len(value), size, value)
|
||||
}
|
||||
}
|
||||
|
||||
if !lr.NextLine() {
|
||||
if err := lr.Err(); err != nil {
|
||||
return fmt.Errorf("cannot read the next log field: %w", err)
|
||||
}
|
||||
|
||||
// add the last log field below before the return
|
||||
}
|
||||
|
||||
if len(name) > maxFieldNameLen {
|
||||
logger.Errorf("%s: field name size should not exceed %d bytes; got %d bytes: %q; skipping this field", streamName, maxFieldNameLen, len(name), name)
|
||||
continue
|
||||
}
|
||||
if !isValidFieldName(name) {
|
||||
logger.Errorf("%s: invalid field name %q; it must consist of `A-Z0-9_` chars and must start from non-digit char; skipping this field", streamName, name)
|
||||
continue
|
||||
}
|
||||
|
||||
if slices.Contains(cp.TimeFields, name) {
|
||||
t, err := strconv.ParseInt(value, 10, 64)
|
||||
// size of binary data encoded as le i64 at the begging
|
||||
idx, err := binary.Decode(data, binary.LittleEndian, &size)
|
||||
if err != nil {
|
||||
logger.Errorf("%s: cannot parse timestamp from the field %q: %w; using the current timestamp", streamName, name, err)
|
||||
ts = 0
|
||||
} else {
|
||||
// Convert journald microsecond timestamp to nanoseconds
|
||||
ts = t * 1e3
|
||||
return fmt.Errorf("failed to extract binary field %q value size: %w", name, err)
|
||||
}
|
||||
// skip binary data size
|
||||
data = data[idx:]
|
||||
if size == 0 {
|
||||
return fmt.Errorf("unexpected zero binary data size decoded %d", size)
|
||||
}
|
||||
if int(size) > len(data) {
|
||||
return fmt.Errorf("binary data size=%d cannot exceed size of the data at buffer=%d", size, len(data))
|
||||
}
|
||||
value = bytesutil.ToUnsafeString(data[:size])
|
||||
data = data[int(size):]
|
||||
// binary data must has new line separator for the new line or next field
|
||||
if len(data) == 0 {
|
||||
return fmt.Errorf("unexpected empty buffer after binary field=%s read", name)
|
||||
}
|
||||
lastB := data[0]
|
||||
if lastB != '\n' {
|
||||
return fmt.Errorf("expected new line separator after binary field=%s, got=%s", name, string(lastB))
|
||||
}
|
||||
data = data[1:]
|
||||
}
|
||||
if len(name) > journaldEntryMaxNameLen {
|
||||
return fmt.Errorf("journald entry name should not exceed %d symbols, got: %q", journaldEntryMaxNameLen, name)
|
||||
}
|
||||
if !allowedJournaldEntryNameChars.MatchString(name) {
|
||||
return fmt.Errorf("journald entry name should consist of `A-Z0-9_` characters and must start from non-digit symbol")
|
||||
}
|
||||
if name == cp.TimeField {
|
||||
n, err := strconv.ParseInt(value, 10, 64)
|
||||
if err != nil {
|
||||
return fmt.Errorf("failed to parse Journald timestamp, %w", err)
|
||||
}
|
||||
ts = n * 1e3
|
||||
continue
|
||||
}
|
||||
|
||||
@@ -322,56 +220,18 @@ func readJournaldLogEntry(streamName string, lr *insertutil.LineReader, lmp inse
|
||||
name = "_msg"
|
||||
}
|
||||
|
||||
if name == "PRIORITY" {
|
||||
priority := journaldPriorityToLevel(value)
|
||||
fb.addField("level", priority)
|
||||
}
|
||||
|
||||
if !strings.HasPrefix(name, "__") || *journaldIncludeEntryMetadata {
|
||||
fb.addField(name, value)
|
||||
if *journaldIncludeEntryMetadata || !strings.HasPrefix(name, "__") {
|
||||
fields = append(fields, logstorage.Field{
|
||||
Name: name,
|
||||
Value: value,
|
||||
})
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func journaldPriorityToLevel(priority string) string {
|
||||
// See https://wiki.archlinux.org/title/Systemd/Journal#Priority_level
|
||||
// and https://grafana.com/docs/grafana/latest/explore/logs-integration/#log-level
|
||||
switch priority {
|
||||
case "0":
|
||||
return "emerg"
|
||||
case "1":
|
||||
return "alert"
|
||||
case "2":
|
||||
return "critical"
|
||||
case "3":
|
||||
return "error"
|
||||
case "4":
|
||||
return "warning"
|
||||
case "5":
|
||||
return "notice"
|
||||
case "6":
|
||||
return "info"
|
||||
case "7":
|
||||
return "debug"
|
||||
default:
|
||||
return priority
|
||||
}
|
||||
}
|
||||
|
||||
func isValidFieldName(s string) bool {
|
||||
if len(s) == 0 {
|
||||
return false
|
||||
}
|
||||
c := s[0]
|
||||
if !(c >= 'A' && c <= 'Z' || c == '_') {
|
||||
return false
|
||||
}
|
||||
|
||||
for i := 1; i < len(s); i++ {
|
||||
c := s[i]
|
||||
if !(c >= 'A' && c <= 'Z' || c >= '0' && c <= '9' || c == '_') {
|
||||
return false
|
||||
if len(fields) > 0 {
|
||||
if ts == 0 {
|
||||
ts = currentTimestamp
|
||||
}
|
||||
lmp.AddRow(ts, fields, nil)
|
||||
}
|
||||
return true
|
||||
return nil
|
||||
}
|
||||
|
||||
@@ -1,81 +1,20 @@
|
||||
package journald
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"net/http"
|
||||
"testing"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutil"
|
||||
)
|
||||
|
||||
func TestIsValidFieldName(t *testing.T) {
|
||||
f := func(name string, resultExpected bool) {
|
||||
t.Helper()
|
||||
|
||||
result := isValidFieldName(name)
|
||||
if result != resultExpected {
|
||||
t.Fatalf("unexpected result for isValidJournaldFieldName(%q); got %v; want %v", name, result, resultExpected)
|
||||
}
|
||||
}
|
||||
|
||||
f("", false)
|
||||
f("a", false)
|
||||
f("1", false)
|
||||
f("_", true)
|
||||
f("X", true)
|
||||
f("Xa", false)
|
||||
f("X_343", true)
|
||||
f("X_0123456789_AZ", true)
|
||||
f("SDDFD sdf", false)
|
||||
}
|
||||
|
||||
func TestGetCommonParams_TimeField(t *testing.T) {
|
||||
f := func(timeFieldHeader, expectedTimeField string) {
|
||||
t.Helper()
|
||||
|
||||
req, err := http.NewRequest("POST", "/insert/journald/upload", nil)
|
||||
if err != nil {
|
||||
t.Fatalf("unexpected error creating request: %s", err)
|
||||
}
|
||||
|
||||
if timeFieldHeader != "" {
|
||||
req.Header.Set("VL-Time-Field", timeFieldHeader)
|
||||
}
|
||||
|
||||
cp, err := getCommonParams(req)
|
||||
if err != nil {
|
||||
t.Fatalf("unexpected error: %s", err)
|
||||
}
|
||||
|
||||
if len(cp.TimeFields) != 1 || cp.TimeFields[0] != expectedTimeField {
|
||||
t.Fatalf("unexpected TimeFields; got %v; want [%s]", cp.TimeFields, expectedTimeField)
|
||||
}
|
||||
}
|
||||
|
||||
// Test default behavior - when no custom time field is specified, journald uses __REALTIME_TIMESTAMP
|
||||
f("", "__REALTIME_TIMESTAMP")
|
||||
|
||||
// Test custom time field - when a custom time field is specified via HTTP header, it's respected
|
||||
f("custom_time", "custom_time")
|
||||
}
|
||||
|
||||
func TestPushJournald_Success(t *testing.T) {
|
||||
func TestPushJournaldOk(t *testing.T) {
|
||||
f := func(src string, timestampsExpected []int64, resultExpected string) {
|
||||
t.Helper()
|
||||
|
||||
tlp := &insertutil.TestLogMessageProcessor{}
|
||||
|
||||
r, err := http.NewRequest("GET", "https://foo.bar/baz", nil)
|
||||
if err != nil {
|
||||
t.Fatalf("cannot create request: %s", err)
|
||||
cp := &insertutil.CommonParams{
|
||||
TimeField: "__REALTIME_TIMESTAMP",
|
||||
MsgFields: []string{"MESSAGE"},
|
||||
}
|
||||
cp, err := getCommonParams(r)
|
||||
if err != nil {
|
||||
t.Fatalf("cannot create commonParams: %s", err)
|
||||
}
|
||||
|
||||
buf := bytes.NewBufferString(src)
|
||||
if err := processStreamInternal("test", buf, tlp, cp); err != nil {
|
||||
if err := parseJournaldRequest([]byte(src), tlp, cp); err != nil {
|
||||
t.Fatalf("unexpected error: %s", err)
|
||||
}
|
||||
|
||||
@@ -83,17 +22,16 @@ func TestPushJournald_Success(t *testing.T) {
|
||||
t.Fatal(err)
|
||||
}
|
||||
}
|
||||
|
||||
// Single event
|
||||
f("__REALTIME_TIMESTAMP=91723819283\nMESSAGE=Test message\n\n",
|
||||
f("__REALTIME_TIMESTAMP=91723819283\nMESSAGE=Test message\n",
|
||||
[]int64{91723819283000},
|
||||
"{\"_msg\":\"Test message\"}",
|
||||
)
|
||||
|
||||
// Multiple events
|
||||
f("__REALTIME_TIMESTAMP=91723819283\nPRIORITY=3\nMESSAGE=Test message\n\n__REALTIME_TIMESTAMP=91723819284\nMESSAGE=Test message2\n",
|
||||
f("__REALTIME_TIMESTAMP=91723819283\nMESSAGE=Test message\n\n__REALTIME_TIMESTAMP=91723819284\nMESSAGE=Test message2\n",
|
||||
[]int64{91723819283000, 91723819284000},
|
||||
"{\"level\":\"error\",\"PRIORITY\":\"3\",\"_msg\":\"Test message\"}\n{\"_msg\":\"Test message2\"}",
|
||||
"{\"_msg\":\"Test message\"}\n{\"_msg\":\"Test message2\"}",
|
||||
)
|
||||
|
||||
// Parse binary data
|
||||
@@ -101,66 +39,30 @@ func TestPushJournald_Success(t *testing.T) {
|
||||
[]int64{1729698775704404000},
|
||||
"{\"E\":\"JobStateChanged\",\"_BOOT_ID\":\"f778b6e2f7584a77b991a2366612a7b5\",\"_UID\":\"0\",\"_GID\":\"0\",\"_MACHINE_ID\":\"a4a970370c30a925df02a13c67167847\",\"_HOSTNAME\":\"ecd5e4555787\",\"_RUNTIME_SCOPE\":\"system\",\"_TRANSPORT\":\"journal\",\"_CAP_EFFECTIVE\":\"1ffffffffff\",\"_SYSTEMD_CGROUP\":\"/init.scope\",\"_SYSTEMD_UNIT\":\"init.scope\",\"_SYSTEMD_SLICE\":\"-.slice\",\"CODE_FILE\":\"\\u003cstdin>\",\"CODE_LINE\":\"1\",\"CODE_FUNC\":\"\\u003cmodule>\",\"SYSLOG_IDENTIFIER\":\"python3\",\"_COMM\":\"python3\",\"_EXE\":\"/usr/bin/python3.12\",\"_CMDLINE\":\"python3\",\"_msg\":\"foo\\nbar\\n\\n\\nasda\\nasda\",\"_PID\":\"2763\",\"_SOURCE_REALTIME_TIMESTAMP\":\"1729698775704375\"}",
|
||||
)
|
||||
|
||||
// Parse binary data with trailing newline
|
||||
f("__REALTIME_TIMESTAMP=1729698775704404\n_CMDLINE=python3\nMESSAGE\n\x14\x00\x00\x00\x00\x00\x00\x00foo\nbar\n\n\nasda\nasda\n\n_PID=2763\n\n",
|
||||
[]int64{1729698775704404000},
|
||||
`{"_CMDLINE":"python3","_msg":"foo\nbar\n\n\nasda\nasda\n","_PID":"2763"}`,
|
||||
)
|
||||
f("__REALTIME_TIMESTAMP=1729698775704404\n_CMDLINE=python3\nMESSAGE\n\x00\x00\x00\x00\x00\x00\x00\x00\n_PID=2763\n\n",
|
||||
[]int64{1729698775704404000},
|
||||
`{"_CMDLINE":"python3","_PID":"2763"}`,
|
||||
)
|
||||
f("__REALTIME_TIMESTAMP=1729698775704404\n_CMDLINE=python3\nMESSAGE\n\x0A\x00\x00\x00\x00\x00\x00\x00123456789\n\n_PID=2763\n\n",
|
||||
[]int64{1729698775704404000},
|
||||
`{"_CMDLINE":"python3","_msg":"123456789\n","_PID":"2763"}`,
|
||||
)
|
||||
f("__REALTIME_TIMESTAMP=1729698775704404\n_CMDLINE=python3\nMESSAGE\n\x0A\x00\x00\x00\x00\x00\x00\x001234567890\n_PID=2763\n\n",
|
||||
[]int64{1729698775704404000},
|
||||
`{"_CMDLINE":"python3","_msg":"1234567890","_PID":"2763"}`,
|
||||
)
|
||||
|
||||
// Empty field name must be ignored
|
||||
f("__REALTIME_TIMESTAMP=91723819283\na=b\n=Test message", nil, "")
|
||||
f("__REALTIME_TIMESTAMP=91723819284\nMESSAGE=Test message2\n\n__REALTIME_TIMESTAMP=91723819283\n=Test message\n", []int64{91723819284000}, `{"_msg":"Test message2"}`)
|
||||
|
||||
// field name starting with number must be ignored
|
||||
f("__REALTIME_TIMESTAMP=91723819283\n1incorrect=Test message\n\n__REALTIME_TIMESTAMP=91723819284\nMESSAGE=Test message2\n\n", []int64{91723819284000}, `{"_msg":"Test message2"}`)
|
||||
|
||||
// field name exceeding 64 bytes limit must be ignored
|
||||
f("__REALTIME_TIMESTAMP=91723819283\ntoolooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooongcorrecooooooooooooong=Test message\n", nil, "")
|
||||
|
||||
// field name with invalid chars must be ignored
|
||||
f("__REALTIME_TIMESTAMP=91723819283\nbadC!@$!@$as=Test message\n", nil, "")
|
||||
}
|
||||
|
||||
func TestPushJournald_Failure(t *testing.T) {
|
||||
f := func(data string) {
|
||||
t.Helper()
|
||||
|
||||
tlp := &insertutil.TestLogMessageProcessor{}
|
||||
|
||||
r, err := http.NewRequest("GET", "https://foo.bar/baz", nil)
|
||||
if err != nil {
|
||||
t.Fatalf("cannot create request: %s", err)
|
||||
cp := &insertutil.CommonParams{
|
||||
TimeField: "__REALTIME_TIMESTAMP",
|
||||
MsgFields: []string{"MESSAGE"},
|
||||
}
|
||||
cp, err := getCommonParams(r)
|
||||
if err != nil {
|
||||
t.Fatalf("cannot create commonParams: %s", err)
|
||||
}
|
||||
|
||||
buf := bytes.NewBufferString(data)
|
||||
if err := processStreamInternal("test", buf, tlp, cp); err == nil {
|
||||
t.Fatalf("expecting non-nil error")
|
||||
if err := parseJournaldRequest([]byte(data), tlp, cp); err == nil {
|
||||
t.Fatalf("expected non nil error")
|
||||
}
|
||||
}
|
||||
|
||||
// too short binary encoded message
|
||||
f("__CURSOR=s=e0afe8412a6a49d2bfcf66aa7927b588;i=1f06;b=f778b6e2f7584a77b991a2366612a7b5;m=300bdfd420;t=62526e1182354;x=930dc44b370963b7\n__REALTIME_TIMESTAMP=1729698775704404\nMESSAGE\n\x13\x00\x00\x00\x00\x00\x00\x00foo\nbar\n\n\nasdaasd")
|
||||
f("__REALTIME_TIMESTAMP=1729698775704404\n_CMDLINE=python3\nMESSAGE\n\x00\x00\x00\x00\x00\x00\x00\x00_PID=2763\n\n")
|
||||
f("__REALTIME_TIMESTAMP=1729698775704404\n_CMDLINE=python3\nMESSAGE\n\x0A\x00\x00\x00\x00\x00\x00\x001234567890_PID=2763\n\n")
|
||||
f("__REALTIME_TIMESTAMP=1729698775704404\n_CMDLINE=python3\nMESSAGE\n\x0A\x00\x00\x00\x00\x00\x00\x00123456789\n_PID=2763\n\n")
|
||||
|
||||
// too long binary encoded message
|
||||
f("__CURSOR=s=e0afe8412a6a49d2bfcf66aa7927b588;i=1f06;b=f778b6e2f7584a77b991a2366612a7b5;m=300bdfd420;t=62526e1182354;x=930dc44b370963b7\n__REALTIME_TIMESTAMP=1729698775704404\nMESSAGE\n\x13\x00\x00\x00\x00\x00\x00\x00foo\nbar\n\n\nasdaasdakljlsfd")
|
||||
// missing new line terminator for binary encoded message
|
||||
f("__CURSOR=s=e0afe8412a6a49d2bfcf66aa7927b588;i=1f06;b=f778b6e2f7584a77b991a2366612a7b5;m=300bdfd420;t=62526e1182354;x=930dc44b370963b7\n__REALTIME_TIMESTAMP=1729698775704404\nMESSAGE\n\x13\x00\x00\x00\x00\x00\x00\x00foo\nbar\n\n\nasdaasda2")
|
||||
// missing new line terminator
|
||||
f("__REALTIME_TIMESTAMP=91723819283\n=Test message")
|
||||
// empty field name
|
||||
f("__REALTIME_TIMESTAMP=91723819283\n=Test message\n")
|
||||
// field name starting with number
|
||||
f("__REALTIME_TIMESTAMP=91723819283\n1incorrect=Test message\n")
|
||||
// field name exceeds 64 limit
|
||||
f("__REALTIME_TIMESTAMP=91723819283\ntoolooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooongcorrecooooooooooooong=Test message\n")
|
||||
// Only allow A-Z0-9 and '_'
|
||||
f("__REALTIME_TIMESTAMP=91723819283\nbadC!@$!@$as=Test message\n")
|
||||
}
|
||||
|
||||
@@ -1,82 +0,0 @@
|
||||
package journald
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"encoding/binary"
|
||||
"fmt"
|
||||
"strings"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutil"
|
||||
)
|
||||
|
||||
func BenchmarkIsValidFieldName(b *testing.B) {
|
||||
b.ReportAllocs()
|
||||
b.SetBytes(int64(len(benchmarkFields)))
|
||||
b.RunParallel(func(pb *testing.PB) {
|
||||
for pb.Next() {
|
||||
for _, field := range benchmarkFields {
|
||||
if !isValidFieldName(field) {
|
||||
panic(fmt.Errorf("cannot validate field %q", field))
|
||||
}
|
||||
}
|
||||
}
|
||||
})
|
||||
}
|
||||
|
||||
var benchmarkFields = strings.Split(
|
||||
"E,_BOOT_ID,_UID,_GID,_MACHINE_ID,_HOSTNAME,_RUNTIME_SCOPE,_TRANSPORT,_CAP_EFFECTIVE,_SYSTEMD_CGROUP,_SYSTEMD_UNIT,"+
|
||||
"_SYSTEMD_SLICE,CODE_FILE,CODE_LINE,CODE_FUNC,SYSLOG_IDENTIFIER,_COMM,_EXE,_CMDLINE,MESSAGE,_PID,_SOURCE_REALTIME_TIMESTAMP,_REALTIME_TIMESTAMP",
|
||||
",")
|
||||
|
||||
func BenchmarkPushJournaldPerformance(b *testing.B) {
|
||||
cp := &insertutil.CommonParams{
|
||||
TimeFields: []string{"__REALTIME_TIMESTAMP"},
|
||||
MsgFields: []string{"MESSAGE"},
|
||||
}
|
||||
const dataChunkSize = 1024 * 1024
|
||||
|
||||
data := generateJournaldData(dataChunkSize)
|
||||
|
||||
b.ReportAllocs()
|
||||
b.SetBytes(int64(len(data)))
|
||||
b.RunParallel(func(pb *testing.PB) {
|
||||
r := &bytes.Reader{}
|
||||
blp := &insertutil.BenchmarkLogMessageProcessor{}
|
||||
for pb.Next() {
|
||||
r.Reset(data)
|
||||
if err := processStreamInternal("performance_test", r, blp, cp); err != nil {
|
||||
panic(fmt.Errorf("unexpected error: %w", err))
|
||||
}
|
||||
}
|
||||
})
|
||||
}
|
||||
|
||||
func generateJournaldData(size int) []byte {
|
||||
var buf []byte
|
||||
timestamp := time.Now().UnixMicro()
|
||||
binaryMsg := []byte("binary message data for performance test")
|
||||
var sizeBuf [8]byte
|
||||
|
||||
for len(buf) < size {
|
||||
timestamp++
|
||||
|
||||
var entry string
|
||||
// Generate a mix of simple and binary messages
|
||||
if timestamp%10 == 0 {
|
||||
// Generate binary message
|
||||
binary.LittleEndian.PutUint64(sizeBuf[:], uint64(len(binaryMsg)))
|
||||
entry = fmt.Sprintf("__REALTIME_TIMESTAMP=%d\nMESSAGE\n%s%s\n\n",
|
||||
timestamp,
|
||||
sizeBuf[:],
|
||||
binaryMsg,
|
||||
)
|
||||
} else {
|
||||
// Generate simple message
|
||||
entry = fmt.Sprintf("__REALTIME_TIMESTAMP=%d\nMESSAGE=Performance test message %d\n\n", timestamp, timestamp)
|
||||
}
|
||||
buf = append(buf, entry...)
|
||||
}
|
||||
return buf
|
||||
}
|
||||
@@ -7,6 +7,7 @@ import (
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
|
||||
@@ -32,7 +33,7 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) {
|
||||
httpserver.Errorf(w, r, "%s", err)
|
||||
return
|
||||
}
|
||||
if err := insertutil.CanWriteData(); err != nil {
|
||||
if err := vlstorage.CanWriteData(); err != nil {
|
||||
httpserver.Errorf(w, r, "%s", err)
|
||||
return
|
||||
}
|
||||
@@ -47,49 +48,34 @@ func RequestHandler(w http.ResponseWriter, r *http.Request) {
|
||||
|
||||
lmp := cp.NewLogMessageProcessor("jsonline", true)
|
||||
streamName := fmt.Sprintf("remoteAddr=%s, requestURI=%q", httpserver.GetQuotedRemoteAddr(r), r.RequestURI)
|
||||
err = processStreamInternal(streamName, reader, cp.TimeFields, cp.MsgFields, lmp)
|
||||
processStreamInternal(streamName, reader, cp.TimeField, cp.MsgFields, lmp)
|
||||
lmp.MustClose()
|
||||
if err != nil {
|
||||
httpserver.Errorf(w, r, "cannot process jsonline request; error: %s", err)
|
||||
return
|
||||
}
|
||||
|
||||
requestDuration.UpdateDuration(startTime)
|
||||
}
|
||||
|
||||
func processStreamInternal(streamName string, r io.Reader, timeFields, msgFields []string, lmp insertutil.LogMessageProcessor) error {
|
||||
func processStreamInternal(streamName string, r io.Reader, timeField string, msgFields []string, lmp insertutil.LogMessageProcessor) {
|
||||
wcr := writeconcurrencylimiter.GetReader(r)
|
||||
defer writeconcurrencylimiter.PutReader(wcr)
|
||||
|
||||
lr := insertutil.NewLineReader(streamName, wcr)
|
||||
|
||||
n := 0
|
||||
errors := 0
|
||||
var lastError error
|
||||
for {
|
||||
ok, err := readLine(lr, timeFields, msgFields, lmp)
|
||||
ok, err := readLine(lr, timeField, msgFields, lmp)
|
||||
wcr.DecConcurrency()
|
||||
if err != nil {
|
||||
lastError = err
|
||||
errors++
|
||||
errorsTotal.Inc()
|
||||
logger.Warnf("jsonline: cannot read line #%d in /jsonline request: %s", n, err)
|
||||
}
|
||||
if !ok {
|
||||
break
|
||||
return
|
||||
}
|
||||
n++
|
||||
}
|
||||
errorsTotal.Add(errors)
|
||||
|
||||
if errors > 0 && n == errors {
|
||||
// Return an error if no logs were processed and there were errors
|
||||
return lastError
|
||||
}
|
||||
|
||||
return nil
|
||||
}
|
||||
|
||||
func readLine(lr *insertutil.LineReader, timeFields, msgFields []string, lmp insertutil.LogMessageProcessor) (bool, error) {
|
||||
func readLine(lr *insertutil.LineReader, timeField string, msgFields []string, lmp insertutil.LogMessageProcessor) (bool, error) {
|
||||
var line []byte
|
||||
for len(line) == 0 {
|
||||
if !lr.NextLine() {
|
||||
@@ -103,11 +89,11 @@ func readLine(lr *insertutil.LineReader, timeFields, msgFields []string, lmp ins
|
||||
defer logstorage.PutJSONParser(p)
|
||||
|
||||
if err := p.ParseLogMessage(line); err != nil {
|
||||
return true, fmt.Errorf("%s; line contents: %q", err, line)
|
||||
return true, fmt.Errorf("cannot parse json-encoded line: %w; line contents: %q", err, line)
|
||||
}
|
||||
ts, err := insertutil.ExtractTimestampFromFields(timeFields, p.Fields)
|
||||
ts, err := insertutil.ExtractTimestampFromFields(timeField, p.Fields)
|
||||
if err != nil {
|
||||
return true, fmt.Errorf("%s; line contents: %q", err, line)
|
||||
return true, fmt.Errorf("cannot get timestamp from json-encoded line: %w; line contents: %q", err, line)
|
||||
}
|
||||
logstorage.RenameField(p.Fields, msgFields, "_msg")
|
||||
lmp.AddRow(ts, p.Fields, nil)
|
||||
@@ -119,5 +105,5 @@ var (
|
||||
requestsTotal = metrics.NewCounter(`vl_http_requests_total{path="/insert/jsonline"}`)
|
||||
errorsTotal = metrics.NewCounter(`vl_http_errors_total{path="/insert/jsonline"}`)
|
||||
|
||||
requestDuration = metrics.NewSummary(`vl_http_request_duration_seconds{path="/insert/jsonline"}`)
|
||||
requestDuration = metrics.NewHistogram(`vl_http_request_duration_seconds{path="/insert/jsonline"}`)
|
||||
)
|
||||
|
||||
@@ -2,23 +2,19 @@ package jsonline
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"strings"
|
||||
"testing"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutil"
|
||||
)
|
||||
|
||||
func TestProcessStreamInternalSuccess(t *testing.T) {
|
||||
func TestProcessStreamInternal(t *testing.T) {
|
||||
f := func(data, timeField, msgField string, timestampsExpected []int64, resultExpected string) {
|
||||
t.Helper()
|
||||
|
||||
timeFields := []string{timeField}
|
||||
msgFields := []string{msgField}
|
||||
tlp := &insertutil.TestLogMessageProcessor{}
|
||||
r := bytes.NewBufferString(data)
|
||||
if err := processStreamInternal("test", r, timeFields, msgFields, tlp); err != nil {
|
||||
t.Fatalf("unexpected error: %s", err)
|
||||
}
|
||||
processStreamInternal("test", r, timeField, msgFields, tlp)
|
||||
|
||||
if err := tlp.Verify(timestampsExpected, resultExpected); err != nil {
|
||||
t.Fatal(err)
|
||||
@@ -48,6 +44,22 @@ func TestProcessStreamInternalSuccess(t *testing.T) {
|
||||
{"message":"baz"}`
|
||||
f(data, timeField, msgField, timestampsExpected, resultExpected)
|
||||
|
||||
// invalid json
|
||||
data = "foobar"
|
||||
timeField = "@timestamp"
|
||||
msgField = "aaa"
|
||||
timestampsExpected = nil
|
||||
resultExpected = ``
|
||||
f(data, timeField, msgField, timestampsExpected, resultExpected)
|
||||
|
||||
// invalid timestamp field
|
||||
data = `{"time":"foobar"}`
|
||||
timeField = "time"
|
||||
msgField = "abc"
|
||||
timestampsExpected = nil
|
||||
resultExpected = ``
|
||||
f(data, timeField, msgField, timestampsExpected, resultExpected)
|
||||
|
||||
// invalid lines among valid lines
|
||||
data = `
|
||||
dsfodmasd
|
||||
@@ -65,33 +77,3 @@ asbsdf
|
||||
{"_msg":"baz"}`
|
||||
f(data, timeField, msgField, timestampsExpected, resultExpected)
|
||||
}
|
||||
|
||||
func TestProcessStreamInternalFailure(t *testing.T) {
|
||||
f := func(data string) {
|
||||
t.Helper()
|
||||
|
||||
tlp := &insertutil.TestLogMessageProcessor{}
|
||||
r := strings.NewReader(data)
|
||||
if err := processStreamInternal("test", r, []string{"time"}, nil, tlp); err == nil {
|
||||
t.Fatalf("expected error, got nil")
|
||||
}
|
||||
|
||||
if err := tlp.Verify(nil, ""); err != nil {
|
||||
t.Fatalf("unexpected error: %s", err)
|
||||
}
|
||||
}
|
||||
|
||||
// invalid json
|
||||
f("foobar")
|
||||
|
||||
f(`foo
|
||||
bar`)
|
||||
|
||||
f(`
|
||||
foo
|
||||
|
||||
`)
|
||||
|
||||
// invalid timestamp field
|
||||
f(`{"time":"foobar"}`)
|
||||
}
|
||||
|
||||
@@ -16,10 +16,10 @@ var disableMessageParsing = flag.Bool("loki.disableMessageParsing", false, "Whet
|
||||
// RequestHandler processes Loki insert requests
|
||||
func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
|
||||
switch path {
|
||||
case "/insert/loki/api/v1/push":
|
||||
case "/api/v1/push":
|
||||
handleInsert(r, w)
|
||||
return true
|
||||
case "/insert/loki/ready":
|
||||
case "/ready":
|
||||
// See https://grafana.com/docs/loki/latest/api/#identify-ready-loki-instance
|
||||
w.WriteHeader(http.StatusOK)
|
||||
w.Write([]byte("ready"))
|
||||
|
||||
@@ -9,6 +9,7 @@ import (
|
||||
"github.com/valyala/fastjson"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
@@ -29,7 +30,7 @@ func handleJSON(r *http.Request, w http.ResponseWriter) {
|
||||
httpserver.Errorf(w, r, "cannot parse common params from request: %s", err)
|
||||
return
|
||||
}
|
||||
if err := insertutil.CanWriteData(); err != nil {
|
||||
if err := vlstorage.CanWriteData(); err != nil {
|
||||
httpserver.Errorf(w, r, "%s", err)
|
||||
return
|
||||
}
|
||||
@@ -58,7 +59,7 @@ func handleJSON(r *http.Request, w http.ResponseWriter) {
|
||||
|
||||
var (
|
||||
requestsJSONTotal = metrics.NewCounter(`vl_http_requests_total{path="/insert/loki/api/v1/push",format="json"}`)
|
||||
requestJSONDuration = metrics.NewSummary(`vl_http_request_duration_seconds{path="/insert/loki/api/v1/push",format="json"}`)
|
||||
requestJSONDuration = metrics.NewHistogram(`vl_http_request_duration_seconds{path="/insert/loki/api/v1/push",format="json"}`)
|
||||
)
|
||||
|
||||
func parseJSONRequest(data []byte, lmp insertutil.LogMessageProcessor, msgFields []string, useDefaultStreamFields, parseMessage bool) error {
|
||||
|
||||
@@ -9,6 +9,7 @@ import (
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
|
||||
@@ -28,7 +29,7 @@ func handleProtobuf(r *http.Request, w http.ResponseWriter) {
|
||||
httpserver.Errorf(w, r, "cannot parse common params from request: %s", err)
|
||||
return
|
||||
}
|
||||
if err := insertutil.CanWriteData(); err != nil {
|
||||
if err := vlstorage.CanWriteData(); err != nil {
|
||||
httpserver.Errorf(w, r, "%s", err)
|
||||
return
|
||||
}
|
||||
@@ -62,7 +63,7 @@ func handleProtobuf(r *http.Request, w http.ResponseWriter) {
|
||||
|
||||
var (
|
||||
requestsProtobufTotal = metrics.NewCounter(`vl_http_requests_total{path="/insert/loki/api/v1/push",format="protobuf"}`)
|
||||
requestProtobufDuration = metrics.NewSummary(`vl_http_request_duration_seconds{path="/insert/loki/api/v1/push",format="protobuf"}`)
|
||||
requestProtobufDuration = metrics.NewHistogram(`vl_http_request_duration_seconds{path="/insert/loki/api/v1/push",format="protobuf"}`)
|
||||
)
|
||||
|
||||
func parseProtobufRequest(data []byte, lmp insertutil.LogMessageProcessor, msgFields []string, useDefaultStreamFields, parseMessage bool) error {
|
||||
|
||||
@@ -1,25 +1,17 @@
|
||||
package vlinsert
|
||||
|
||||
import (
|
||||
"flag"
|
||||
"fmt"
|
||||
"net/http"
|
||||
"strings"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/datadog"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/elasticsearch"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/internalinsert"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/journald"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/jsonline"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/loki"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/opentelemetry"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/syslog"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
)
|
||||
|
||||
var (
|
||||
disableInsert = flag.Bool("insert.disable", false, "Whether to disable /insert/* HTTP endpoints")
|
||||
disableInternal = flag.Bool("internalinsert.disable", false, "Whether to disable /internal/insert HTTP endpoint")
|
||||
)
|
||||
|
||||
// Init initializes vlinsert
|
||||
@@ -34,55 +26,44 @@ func Stop() {
|
||||
|
||||
// RequestHandler handles insert requests for VictoriaLogs
|
||||
func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
|
||||
path := strings.ReplaceAll(r.URL.Path, "//", "/")
|
||||
path := r.URL.Path
|
||||
|
||||
if strings.HasPrefix(path, "/insert/") {
|
||||
if *disableInsert {
|
||||
httpserver.Errorf(w, r, "requests to /insert/* are disabled with -insert.disable command-line flag")
|
||||
return true
|
||||
}
|
||||
|
||||
return insertHandler(w, r, path)
|
||||
if !strings.HasPrefix(path, "/insert/") {
|
||||
// Skip requests, which do not start with /insert/, since these aren't our requests.
|
||||
return false
|
||||
}
|
||||
path = strings.TrimPrefix(path, "/insert")
|
||||
path = strings.ReplaceAll(path, "//", "/")
|
||||
|
||||
if path == "/internal/insert" {
|
||||
if *disableInternal || *disableInsert {
|
||||
httpserver.Errorf(w, r, "requests to /internal/insert are disabled with -internalinsert.disable or -insert.disable command-line flag")
|
||||
return true
|
||||
}
|
||||
internalinsert.RequestHandler(w, r)
|
||||
return true
|
||||
}
|
||||
|
||||
return false
|
||||
}
|
||||
|
||||
func insertHandler(w http.ResponseWriter, r *http.Request, path string) bool {
|
||||
switch path {
|
||||
case "/insert/jsonline":
|
||||
case "/jsonline":
|
||||
jsonline.RequestHandler(w, r)
|
||||
return true
|
||||
case "/insert/ready":
|
||||
case "/ready":
|
||||
w.Header().Set("Content-Type", "application/json")
|
||||
w.WriteHeader(200)
|
||||
fmt.Fprintf(w, `{"status":"ok"}`)
|
||||
return true
|
||||
}
|
||||
switch {
|
||||
// some clients may omit trailing slash at elasticsearch protocol.
|
||||
// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/8353
|
||||
case strings.HasPrefix(path, "/insert/elasticsearch"):
|
||||
case strings.HasPrefix(path, "/elasticsearch"):
|
||||
// some clients may omit trailing slash
|
||||
// see https://github.com/VictoriaMetrics/VictoriaMetrics/issues/8353
|
||||
path = strings.TrimPrefix(path, "/elasticsearch")
|
||||
return elasticsearch.RequestHandler(path, w, r)
|
||||
|
||||
case strings.HasPrefix(path, "/insert/loki/"):
|
||||
case strings.HasPrefix(path, "/loki/"):
|
||||
path = strings.TrimPrefix(path, "/loki")
|
||||
return loki.RequestHandler(path, w, r)
|
||||
case strings.HasPrefix(path, "/insert/opentelemetry/"):
|
||||
case strings.HasPrefix(path, "/opentelemetry/"):
|
||||
path = strings.TrimPrefix(path, "/opentelemetry")
|
||||
return opentelemetry.RequestHandler(path, w, r)
|
||||
case strings.HasPrefix(path, "/insert/journald/"):
|
||||
case strings.HasPrefix(path, "/journald/"):
|
||||
path = strings.TrimPrefix(path, "/journald")
|
||||
return journald.RequestHandler(path, w, r)
|
||||
case strings.HasPrefix(path, "/insert/datadog/"):
|
||||
case strings.HasPrefix(path, "/datadog/"):
|
||||
path = strings.TrimPrefix(path, "/datadog")
|
||||
return datadog.RequestHandler(path, w, r)
|
||||
default:
|
||||
return false
|
||||
}
|
||||
|
||||
return false
|
||||
}
|
||||
|
||||
@@ -6,11 +6,13 @@ import (
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/opentelemetry/pb"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/slicesutil"
|
||||
"github.com/VictoriaMetrics/metrics"
|
||||
)
|
||||
|
||||
@@ -21,7 +23,7 @@ func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
|
||||
switch path {
|
||||
// use the same path as opentelemetry collector
|
||||
// https://opentelemetry.io/docs/specs/otlp/#otlphttp-request
|
||||
case "/insert/opentelemetry/v1/logs":
|
||||
case "/v1/logs":
|
||||
if r.Header.Get("Content-Type") == "application/json" {
|
||||
httpserver.Errorf(w, r, "json encoding isn't supported for opentelemetry format. Use protobuf encoding")
|
||||
return true
|
||||
@@ -42,7 +44,7 @@ func handleProtobuf(r *http.Request, w http.ResponseWriter) {
|
||||
httpserver.Errorf(w, r, "cannot parse common params from request: %s", err)
|
||||
return
|
||||
}
|
||||
if err := insertutil.CanWriteData(); err != nil {
|
||||
if err := vlstorage.CanWriteData(); err != nil {
|
||||
httpserver.Errorf(w, r, "%s", err)
|
||||
return
|
||||
}
|
||||
@@ -51,7 +53,7 @@ func handleProtobuf(r *http.Request, w http.ResponseWriter) {
|
||||
err = protoparserutil.ReadUncompressedData(r.Body, encoding, maxRequestSize, func(data []byte) error {
|
||||
lmp := cp.NewLogMessageProcessor("opentelelemtry_protobuf", false)
|
||||
useDefaultStreamFields := len(cp.StreamFields) == 0
|
||||
err := pushProtobufRequest(data, lmp, cp.MsgFields, useDefaultStreamFields)
|
||||
err := pushProtobufRequest(data, lmp, useDefaultStreamFields)
|
||||
lmp.MustClose()
|
||||
return err
|
||||
})
|
||||
@@ -70,10 +72,10 @@ var (
|
||||
requestsProtobufTotal = metrics.NewCounter(`vl_http_requests_total{path="/insert/opentelemetry/v1/logs",format="protobuf"}`)
|
||||
errorsTotal = metrics.NewCounter(`vl_http_errors_total{path="/insert/opentelemetry/v1/logs",format="protobuf"}`)
|
||||
|
||||
requestProtobufDuration = metrics.NewSummary(`vl_http_request_duration_seconds{path="/insert/opentelemetry/v1/logs",format="protobuf"}`)
|
||||
requestProtobufDuration = metrics.NewHistogram(`vl_http_request_duration_seconds{path="/insert/opentelemetry/v1/logs",format="protobuf"}`)
|
||||
)
|
||||
|
||||
func pushProtobufRequest(data []byte, lmp insertutil.LogMessageProcessor, msgFields []string, useDefaultStreamFields bool) error {
|
||||
func pushProtobufRequest(data []byte, lmp insertutil.LogMessageProcessor, useDefaultStreamFields bool) error {
|
||||
var req pb.ExportLogsServiceRequest
|
||||
if err := req.UnmarshalProtobuf(data); err != nil {
|
||||
errorsTotal.Inc()
|
||||
@@ -82,31 +84,35 @@ func pushProtobufRequest(data []byte, lmp insertutil.LogMessageProcessor, msgFie
|
||||
|
||||
var commonFields []logstorage.Field
|
||||
for _, rl := range req.ResourceLogs {
|
||||
commonFields = commonFields[:0]
|
||||
commonFields = appendKeyValues(commonFields, rl.Resource.Attributes, "")
|
||||
attributes := rl.Resource.Attributes
|
||||
commonFields = slicesutil.SetLength(commonFields, len(attributes))
|
||||
for i, attr := range attributes {
|
||||
commonFields[i].Name = attr.Key
|
||||
commonFields[i].Value = attr.Value.FormatString(true)
|
||||
}
|
||||
commonFieldsLen := len(commonFields)
|
||||
for _, sc := range rl.ScopeLogs {
|
||||
commonFields = pushFieldsFromScopeLogs(&sc, commonFields[:commonFieldsLen], lmp, msgFields, useDefaultStreamFields)
|
||||
commonFields = pushFieldsFromScopeLogs(&sc, commonFields[:commonFieldsLen], lmp, useDefaultStreamFields)
|
||||
}
|
||||
}
|
||||
|
||||
return nil
|
||||
}
|
||||
|
||||
func pushFieldsFromScopeLogs(sc *pb.ScopeLogs, commonFields []logstorage.Field, lmp insertutil.LogMessageProcessor, msgFields []string, useDefaultStreamFields bool) []logstorage.Field {
|
||||
func pushFieldsFromScopeLogs(sc *pb.ScopeLogs, commonFields []logstorage.Field, lmp insertutil.LogMessageProcessor, useDefaultStreamFields bool) []logstorage.Field {
|
||||
fields := commonFields
|
||||
for _, lr := range sc.LogRecords {
|
||||
fields = fields[:len(commonFields)]
|
||||
if lr.Body.KeyValueList != nil {
|
||||
fields = appendKeyValues(fields, lr.Body.KeyValueList.Values, "")
|
||||
logstorage.RenameField(fields[len(commonFields):], msgFields, "_msg")
|
||||
} else {
|
||||
fields = append(fields, logstorage.Field{
|
||||
Name: "_msg",
|
||||
Value: lr.Body.FormatString(true),
|
||||
})
|
||||
for _, attr := range lr.Attributes {
|
||||
fields = append(fields, logstorage.Field{
|
||||
Name: "_msg",
|
||||
Value: lr.Body.FormatString(true),
|
||||
Name: attr.Key,
|
||||
Value: attr.Value.FormatString(true),
|
||||
})
|
||||
}
|
||||
fields = appendKeyValues(fields, lr.Attributes, "")
|
||||
if len(lr.TraceID) > 0 {
|
||||
fields = append(fields, logstorage.Field{
|
||||
Name: "trace_id",
|
||||
@@ -132,22 +138,3 @@ func pushFieldsFromScopeLogs(sc *pb.ScopeLogs, commonFields []logstorage.Field,
|
||||
}
|
||||
return fields
|
||||
}
|
||||
|
||||
func appendKeyValues(fields []logstorage.Field, kvs []*pb.KeyValue, parentField string) []logstorage.Field {
|
||||
for _, attr := range kvs {
|
||||
fieldName := attr.Key
|
||||
if parentField != "" {
|
||||
fieldName = parentField + "." + fieldName
|
||||
}
|
||||
|
||||
if attr.Value.KeyValueList != nil {
|
||||
fields = appendKeyValues(fields, attr.Value.KeyValueList.Values, fieldName)
|
||||
} else {
|
||||
fields = append(fields, logstorage.Field{
|
||||
Name: fieldName,
|
||||
Value: attr.Value.FormatString(true),
|
||||
})
|
||||
}
|
||||
}
|
||||
return fields
|
||||
}
|
||||
|
||||
@@ -16,7 +16,7 @@ func TestPushProtoOk(t *testing.T) {
|
||||
|
||||
pData := lr.MarshalProtobuf(nil)
|
||||
tlp := &insertutil.TestLogMessageProcessor{}
|
||||
if err := pushProtobufRequest(pData, tlp, nil, false); err != nil {
|
||||
if err := pushProtobufRequest(pData, tlp, false); err != nil {
|
||||
t.Fatalf("unexpected error: %s", err)
|
||||
}
|
||||
|
||||
@@ -41,26 +41,6 @@ func TestPushProtoOk(t *testing.T) {
|
||||
`{"_msg":"log-line-message","severity":"Trace"}`,
|
||||
)
|
||||
|
||||
// severities mapping
|
||||
f([]pb.ResourceLogs{
|
||||
{
|
||||
ScopeLogs: []pb.ScopeLogs{
|
||||
{
|
||||
LogRecords: []pb.LogRecord{
|
||||
{Attributes: []*pb.KeyValue{}, TimeUnixNano: 1234, SeverityNumber: 1, Body: pb.AnyValue{StringValue: ptrTo("log-line-message")}},
|
||||
{Attributes: []*pb.KeyValue{}, TimeUnixNano: 1234, SeverityNumber: 13, Body: pb.AnyValue{StringValue: ptrTo("log-line-message")}},
|
||||
{Attributes: []*pb.KeyValue{}, TimeUnixNano: 1234, SeverityNumber: 24, Body: pb.AnyValue{StringValue: ptrTo("log-line-message")}},
|
||||
},
|
||||
},
|
||||
},
|
||||
},
|
||||
},
|
||||
[]int64{1234, 1234, 1234},
|
||||
`{"_msg":"log-line-message","severity":"Trace"}
|
||||
{"_msg":"log-line-message","severity":"Warn"}
|
||||
{"_msg":"log-line-message","severity":"Fatal4"}`,
|
||||
)
|
||||
|
||||
// multi-line with resource attributes
|
||||
f([]pb.ResourceLogs{
|
||||
{
|
||||
@@ -80,7 +60,7 @@ func TestPushProtoOk(t *testing.T) {
|
||||
{
|
||||
LogRecords: []pb.LogRecord{
|
||||
{Attributes: []*pb.KeyValue{}, TimeUnixNano: 1234, SeverityNumber: 1, Body: pb.AnyValue{StringValue: ptrTo("log-line-message")}},
|
||||
{Attributes: []*pb.KeyValue{}, TimeUnixNano: 1235, SeverityNumber: 25, Body: pb.AnyValue{StringValue: ptrTo("log-line-message-msg-2")}},
|
||||
{Attributes: []*pb.KeyValue{}, TimeUnixNano: 1235, SeverityNumber: 21, Body: pb.AnyValue{StringValue: ptrTo("log-line-message-msg-2")}},
|
||||
{Attributes: []*pb.KeyValue{}, TimeUnixNano: 1236, SeverityNumber: -1, Body: pb.AnyValue{StringValue: ptrTo("log-line-message-msg-2")}},
|
||||
},
|
||||
},
|
||||
@@ -88,9 +68,9 @@ func TestPushProtoOk(t *testing.T) {
|
||||
},
|
||||
},
|
||||
[]int64{1234, 1235, 1236},
|
||||
`{"logger":"context","instance_id":"10","node_taints.role":"dev","node_taints.cluster_load_percent":"0.55","_msg":"log-line-message","severity":"Trace"}
|
||||
{"logger":"context","instance_id":"10","node_taints.role":"dev","node_taints.cluster_load_percent":"0.55","_msg":"log-line-message-msg-2","severity":"Unspecified"}
|
||||
{"logger":"context","instance_id":"10","node_taints.role":"dev","node_taints.cluster_load_percent":"0.55","_msg":"log-line-message-msg-2","severity":"Unspecified"}`,
|
||||
`{"logger":"context","instance_id":"10","node_taints":"{\"role\":\"dev\",\"cluster_load_percent\":0.55}","_msg":"log-line-message","severity":"Trace"}
|
||||
{"logger":"context","instance_id":"10","node_taints":"{\"role\":\"dev\",\"cluster_load_percent\":0.55}","_msg":"log-line-message-msg-2","severity":"Unspecified"}
|
||||
{"logger":"context","instance_id":"10","node_taints":"{\"role\":\"dev\",\"cluster_load_percent\":0.55}","_msg":"log-line-message-msg-2","severity":"Unspecified"}`,
|
||||
)
|
||||
|
||||
// multi-scope with resource attributes and multi-line
|
||||
@@ -136,71 +116,14 @@ func TestPushProtoOk(t *testing.T) {
|
||||
},
|
||||
},
|
||||
[]int64{1234, 1235, 2345, 2346, 2347, 2348, 3333},
|
||||
`{"logger":"context","instance_id":"10","node_taints.role":"dev","node_taints.cluster_load_percent":"0.55","_msg":"log-line-message","severity":"Trace"}
|
||||
{"logger":"context","instance_id":"10","node_taints.role":"dev","node_taints.cluster_load_percent":"0.55","_msg":"log-line-message-msg-2","severity":"Debug"}
|
||||
`{"logger":"context","instance_id":"10","node_taints":"{\"role\":\"dev\",\"cluster_load_percent\":0.55}","_msg":"log-line-message","severity":"Trace"}
|
||||
{"logger":"context","instance_id":"10","node_taints":"{\"role\":\"dev\",\"cluster_load_percent\":0.55}","_msg":"log-line-message-msg-2","severity":"Debug"}
|
||||
{"_msg":"log-line-resource-scope-1-0-0","severity":"Info2"}
|
||||
{"_msg":"log-line-resource-scope-1-0-1","severity":"Info2"}
|
||||
{"_msg":"log-line-resource-scope-1-1-0","severity":"Info4"}
|
||||
{"_msg":"log-line-resource-scope-1-1-1","trace_id":"1234","span_id":"45","severity":"Info4"}
|
||||
{"_msg":"log-line-resource-scope-1-1-2","trace_id":"4bf92f3577b34da6a3ce929d0e0e4736","span_id":"00f067aa0ba902b7","severity":"Unspecified"}`,
|
||||
)
|
||||
|
||||
// nested fields
|
||||
f([]pb.ResourceLogs{
|
||||
{
|
||||
ScopeLogs: []pb.ScopeLogs{
|
||||
{
|
||||
LogRecords: []pb.LogRecord{
|
||||
{
|
||||
TimeUnixNano: 1234,
|
||||
Body: pb.AnyValue{StringValue: ptrTo("nested fields")},
|
||||
Attributes: []*pb.KeyValue{
|
||||
{Key: "error", Value: &pb.AnyValue{KeyValueList: &pb.KeyValueList{Values: []*pb.KeyValue{
|
||||
{
|
||||
Key: "type",
|
||||
Value: &pb.AnyValue{StringValue: ptrTo("document_parsing_exception")},
|
||||
},
|
||||
{
|
||||
Key: "reason",
|
||||
Value: &pb.AnyValue{StringValue: ptrTo("failed to parse field [_msg] of type [text]")},
|
||||
},
|
||||
{
|
||||
Key: "caused_by",
|
||||
Value: &pb.AnyValue{KeyValueList: &pb.KeyValueList{Values: []*pb.KeyValue{
|
||||
{
|
||||
Key: "type",
|
||||
Value: &pb.AnyValue{StringValue: ptrTo("x_content_parse_exception")},
|
||||
},
|
||||
{
|
||||
Key: "reason",
|
||||
Value: &pb.AnyValue{StringValue: ptrTo("unexpected end-of-input in VALUE_STRING")},
|
||||
},
|
||||
{
|
||||
Key: "caused_by",
|
||||
Value: &pb.AnyValue{KeyValueList: &pb.KeyValueList{Values: []*pb.KeyValue{
|
||||
{
|
||||
Key: "type",
|
||||
Value: &pb.AnyValue{StringValue: ptrTo("json_e_o_f_exception")},
|
||||
},
|
||||
{
|
||||
Key: "reason",
|
||||
Value: &pb.AnyValue{StringValue: ptrTo("eof")},
|
||||
},
|
||||
}}},
|
||||
},
|
||||
}}},
|
||||
},
|
||||
}}}},
|
||||
},
|
||||
},
|
||||
},
|
||||
},
|
||||
},
|
||||
},
|
||||
}, []int64{1234},
|
||||
`{"_msg":"nested fields","error.type":"document_parsing_exception","error.reason":"failed to parse field [_msg] of type [text]",`+
|
||||
`"error.caused_by.type":"x_content_parse_exception","error.caused_by.reason":"unexpected end-of-input in VALUE_STRING",`+
|
||||
`"error.caused_by.caused_by.type":"json_e_o_f_exception","error.caused_by.caused_by.reason":"eof","severity":"Unspecified"}`)
|
||||
}
|
||||
|
||||
func ptrTo[T any](s T) *T {
|
||||
|
||||
@@ -27,7 +27,7 @@ func benchmarkParseProtobufRequest(b *testing.B, streams, rows, labels int) {
|
||||
b.RunParallel(func(pb *testing.PB) {
|
||||
body := getProtobufBody(streams, rows, labels)
|
||||
for pb.Next() {
|
||||
if err := pushProtobufRequest(body, blp, nil, false); err != nil {
|
||||
if err := pushProtobufRequest(body, blp, false); err != nil {
|
||||
panic(fmt.Errorf("unexpected error: %w", err))
|
||||
}
|
||||
}
|
||||
|
||||
@@ -17,6 +17,7 @@ import (
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/cgroup"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
|
||||
@@ -44,11 +45,6 @@ var (
|
||||
ignoreFieldsUDP = flagutil.NewArrayString("syslog.ignoreFields.udp", "Fields to ignore at logs ingested via the corresponding -syslog.listenAddr.udp. "+
|
||||
`See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#dropping-fields`)
|
||||
|
||||
decolorizeFieldsTCP = flagutil.NewArrayString("syslog.decolorizeFields.tcp", "Fields to remove ANSI color codes across logs ingested via the corresponding -syslog.listenAddr.tcp. "+
|
||||
`See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#decolorizing-fields`)
|
||||
decolorizeFieldsUDP = flagutil.NewArrayString("syslog.decolorizeFields.udp", "Fields to remove ANSI color codes across logs ingested via the corresponding -syslog.listenAddr.udp. "+
|
||||
`See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#decolorizing-fields`)
|
||||
|
||||
extraFieldsTCP = flagutil.NewArrayString("syslog.extraFields.tcp", "Fields to add to logs ingested via the corresponding -syslog.listenAddr.tcp. "+
|
||||
`See https://docs.victoriametrics.com/victorialogs/data-ingestion/syslog/#adding-extra-fields`)
|
||||
extraFieldsUDP = flagutil.NewArrayString("syslog.extraFields.udp", "Fields to add to logs ingested via the corresponding -syslog.listenAddr.udp. "+
|
||||
@@ -192,12 +188,6 @@ func runUDPListener(addr string, argIdx int) {
|
||||
logger.Fatalf("cannot parse -syslog.ignoreFields.udp=%q for -syslog.listenAddr.udp=%q: %s", ignoreFieldsStr, addr, err)
|
||||
}
|
||||
|
||||
decolorizeFieldsStr := decolorizeFieldsUDP.GetOptionalArg(argIdx)
|
||||
decolorizeFields, err := parseFieldsList(decolorizeFieldsStr)
|
||||
if err != nil {
|
||||
logger.Fatalf("cannot parse -syslog.decolorizeFields.udp=%q for -syslog.listenAddr.udp=%q: %s", decolorizeFieldsStr, addr, err)
|
||||
}
|
||||
|
||||
extraFieldsStr := extraFieldsUDP.GetOptionalArg(argIdx)
|
||||
extraFields, err := parseExtraFields(extraFieldsStr)
|
||||
if err != nil {
|
||||
@@ -206,7 +196,7 @@ func runUDPListener(addr string, argIdx int) {
|
||||
|
||||
doneCh := make(chan struct{})
|
||||
go func() {
|
||||
serveUDP(ln, tenantID, compressMethod, useLocalTimestamp, streamFields, ignoreFields, decolorizeFields, extraFields)
|
||||
serveUDP(ln, tenantID, compressMethod, useLocalTimestamp, streamFields, ignoreFields, extraFields)
|
||||
close(doneCh)
|
||||
}()
|
||||
|
||||
@@ -259,12 +249,6 @@ func runTCPListener(addr string, argIdx int) {
|
||||
logger.Fatalf("cannot parse -syslog.ignoreFields.tcp=%q for -syslog.listenAddr.tcp=%q: %s", ignoreFieldsStr, addr, err)
|
||||
}
|
||||
|
||||
decolorizeFieldsStr := decolorizeFieldsTCP.GetOptionalArg(argIdx)
|
||||
decolorizeFields, err := parseFieldsList(decolorizeFieldsStr)
|
||||
if err != nil {
|
||||
logger.Fatalf("cannot parse -syslog.decolorizeFields.tcp=%q for -syslog.listenAddr.tcp=%q: %s", decolorizeFieldsStr, addr, err)
|
||||
}
|
||||
|
||||
extraFieldsStr := extraFieldsTCP.GetOptionalArg(argIdx)
|
||||
extraFields, err := parseExtraFields(extraFieldsStr)
|
||||
if err != nil {
|
||||
@@ -273,7 +257,7 @@ func runTCPListener(addr string, argIdx int) {
|
||||
|
||||
doneCh := make(chan struct{})
|
||||
go func() {
|
||||
serveTCP(ln, tenantID, compressMethod, useLocalTimestamp, streamFields, ignoreFields, decolorizeFields, extraFields)
|
||||
serveTCP(ln, tenantID, compressMethod, useLocalTimestamp, streamFields, ignoreFields, extraFields)
|
||||
close(doneCh)
|
||||
}()
|
||||
|
||||
@@ -295,7 +279,7 @@ func checkCompressMethod(compressMethod, addr, protocol string) {
|
||||
}
|
||||
}
|
||||
|
||||
func serveUDP(ln net.PacketConn, tenantID logstorage.TenantID, encoding string, useLocalTimestamp bool, streamFields, ignoreFields, decolorizeFields []string, extraFields []logstorage.Field) {
|
||||
func serveUDP(ln net.PacketConn, tenantID logstorage.TenantID, encoding string, useLocalTimestamp bool, streamFields, ignoreFields []string, extraFields []logstorage.Field) {
|
||||
gomaxprocs := cgroup.AvailableCPUs()
|
||||
var wg sync.WaitGroup
|
||||
localAddr := ln.LocalAddr()
|
||||
@@ -303,7 +287,7 @@ func serveUDP(ln net.PacketConn, tenantID logstorage.TenantID, encoding string,
|
||||
wg.Add(1)
|
||||
go func() {
|
||||
defer wg.Done()
|
||||
cp := insertutil.GetCommonParamsForSyslog(tenantID, streamFields, ignoreFields, decolorizeFields, extraFields)
|
||||
cp := insertutil.GetCommonParamsForSyslog(tenantID, streamFields, ignoreFields, extraFields)
|
||||
var bb bytesutil.ByteBuffer
|
||||
bb.B = bytesutil.ResizeNoCopyNoOverallocate(bb.B, 64*1024)
|
||||
for {
|
||||
@@ -337,7 +321,7 @@ func serveUDP(ln net.PacketConn, tenantID logstorage.TenantID, encoding string,
|
||||
wg.Wait()
|
||||
}
|
||||
|
||||
func serveTCP(ln net.Listener, tenantID logstorage.TenantID, encoding string, useLocalTimestamp bool, streamFields, ignoreFields, decolorizeFields []string, extraFields []logstorage.Field) {
|
||||
func serveTCP(ln net.Listener, tenantID logstorage.TenantID, encoding string, useLocalTimestamp bool, streamFields, ignoreFields []string, extraFields []logstorage.Field) {
|
||||
var cm ingestserver.ConnsMap
|
||||
cm.Init("syslog")
|
||||
|
||||
@@ -367,7 +351,7 @@ func serveTCP(ln net.Listener, tenantID logstorage.TenantID, encoding string, us
|
||||
|
||||
wg.Add(1)
|
||||
go func() {
|
||||
cp := insertutil.GetCommonParamsForSyslog(tenantID, streamFields, ignoreFields, decolorizeFields, extraFields)
|
||||
cp := insertutil.GetCommonParamsForSyslog(tenantID, streamFields, ignoreFields, extraFields)
|
||||
if err := processStream("tcp", c, encoding, useLocalTimestamp, cp); err != nil {
|
||||
logger.Errorf("syslog: cannot process TCP data at %q: %s", addr, err)
|
||||
}
|
||||
@@ -384,7 +368,7 @@ func serveTCP(ln net.Listener, tenantID logstorage.TenantID, encoding string, us
|
||||
|
||||
// processStream parses a stream of syslog messages from r and ingests them into vlstorage.
|
||||
func processStream(protocol string, r io.Reader, encoding string, useLocalTimestamp bool, cp *insertutil.CommonParams) error {
|
||||
if err := insertutil.CanWriteData(); err != nil {
|
||||
if err := vlstorage.CanWriteData(); err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
@@ -551,7 +535,7 @@ func processLine(line []byte, currentYear int, timezone *time.Location, useLocal
|
||||
if useLocalTimestamp {
|
||||
ts = time.Now().UnixNano()
|
||||
} else {
|
||||
nsecs, err := insertutil.ExtractTimestampFromFields(timeFields, p.Fields)
|
||||
nsecs, err := insertutil.ExtractTimestampFromFields("timestamp", p.Fields)
|
||||
if err != nil {
|
||||
return fmt.Errorf("cannot get timestamp from syslog line %q: %w", line, err)
|
||||
}
|
||||
@@ -564,7 +548,6 @@ func processLine(line []byte, currentYear int, timezone *time.Location, useLocal
|
||||
return nil
|
||||
}
|
||||
|
||||
var timeFields = []string{"timestamp"}
|
||||
var msgFields = []string{"message"}
|
||||
|
||||
var (
|
||||
|
||||
@@ -101,8 +101,8 @@ func TestProcessStreamInternal_Success(t *testing.T) {
|
||||
currentYear := 2023
|
||||
timestampsExpected := []int64{1685794113000000000, 1685880513000000000, 1685814132345000000}
|
||||
resultExpected := `{"format":"rfc3164","hostname":"abcd","app_name":"systemd","_msg":"Starting Update the local ESM caches..."}
|
||||
{"priority":"165","facility_keyword":"local4","level":"notice","facility":"20","severity":"5","format":"rfc3164","hostname":"abcd","app_name":"systemd","proc_id":"345","_msg":"abc defg"}
|
||||
{"priority":"123","facility_keyword":"solaris-cron","level":"error","facility":"15","severity":"3","format":"rfc5424","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":"165","facility":"20","severity":"5","format":"rfc3164","hostname":"abcd","app_name":"systemd","proc_id":"345","_msg":"abc defg"}
|
||||
{"priority":"123","facility":"15","severity":"3","format":"rfc5424","hostname":"mymachine.example.com","app_name":"appname","proc_id":"12345","msg_id":"ID47","exampleSDID@32473.iut":"3","exampleSDID@32473.eventSource":"Application 123 = ] 56","exampleSDID@32473.eventID":"11211","_msg":"This is a test message with structured data."}`
|
||||
f(data, currentYear, timestampsExpected, resultExpected)
|
||||
}
|
||||
|
||||
|
||||
@@ -17,7 +17,7 @@ func isTerminal() bool {
|
||||
return isatty.IsTerminal(os.Stdout.Fd()) && isatty.IsTerminal(os.Stderr.Fd())
|
||||
}
|
||||
|
||||
func readWithLess(r io.Reader, disableColors, wrapLongLines bool) error {
|
||||
func readWithLess(r io.Reader, wrapLongLines bool) error {
|
||||
if !isTerminal() {
|
||||
// Just write everything to stdout if no terminal is available.
|
||||
_, err := io.Copy(os.Stdout, r)
|
||||
@@ -49,9 +49,6 @@ func readWithLess(r io.Reader, disableColors, wrapLongLines bool) error {
|
||||
return fmt.Errorf("cannot find 'less' command: %w", err)
|
||||
}
|
||||
opts := []string{"less", "-F", "-X"}
|
||||
if !disableColors {
|
||||
opts = append(opts, "-R")
|
||||
}
|
||||
if !wrapLongLines {
|
||||
opts = append(opts, "-S")
|
||||
}
|
||||
|
||||
@@ -91,7 +91,6 @@ func runReadlineLoop(rl *readline.Instance, incompleteLine *string) {
|
||||
}
|
||||
|
||||
outputMode := outputModeJSONMultiline
|
||||
disableColors := true
|
||||
wrapLongLines := false
|
||||
s := ""
|
||||
for {
|
||||
@@ -101,7 +100,7 @@ func runReadlineLoop(rl *readline.Instance, incompleteLine *string) {
|
||||
case io.EOF:
|
||||
if s != "" {
|
||||
// This is non-interactive query execution.
|
||||
executeQuery(context.Background(), rl, s, outputMode, disableColors, wrapLongLines)
|
||||
executeQuery(context.Background(), rl, s, outputMode, wrapLongLines)
|
||||
}
|
||||
return
|
||||
case readline.ErrInterrupt:
|
||||
@@ -177,24 +176,6 @@ func runReadlineLoop(rl *readline.Instance, incompleteLine *string) {
|
||||
s = ""
|
||||
continue
|
||||
}
|
||||
if s == `\disable_colors` {
|
||||
if !disableColors {
|
||||
disableColors = true
|
||||
fmt.Fprintf(rl, `disabled colors in compact output mode; enter \enable_colors for enabling it`+"\n")
|
||||
}
|
||||
historyLines = pushToHistory(rl, historyLines, s)
|
||||
s = ""
|
||||
continue
|
||||
}
|
||||
if s == `\enable_colors` {
|
||||
if disableColors {
|
||||
disableColors = false
|
||||
fmt.Fprintf(rl, `enabled colors in compact output mode; type \disable_colors for disabling it`+"\n")
|
||||
}
|
||||
historyLines = pushToHistory(rl, historyLines, s)
|
||||
s = ""
|
||||
continue
|
||||
}
|
||||
if line != "" && !strings.HasSuffix(line, ";") {
|
||||
// Assume the query is incomplete and allow the user finishing the query on the next line
|
||||
s += "\n"
|
||||
@@ -204,7 +185,7 @@ func runReadlineLoop(rl *readline.Instance, incompleteLine *string) {
|
||||
|
||||
// Execute the query
|
||||
ctx, cancel := signal.NotifyContext(context.Background(), os.Interrupt)
|
||||
executeQuery(ctx, rl, s, outputMode, disableColors, wrapLongLines)
|
||||
executeQuery(ctx, rl, s, outputMode, wrapLongLines)
|
||||
cancel()
|
||||
|
||||
historyLines = pushToHistory(rl, historyLines, s)
|
||||
@@ -292,15 +273,13 @@ func printCommandsHelp(w io.Writer) {
|
||||
\c - compact output mode
|
||||
\logfmt - logfmt output mode
|
||||
\wrap_long_lines - toggles wrapping long lines
|
||||
\enable_colors - enable ANSI colors in compact output mode
|
||||
\disable_colors - disable ANSI colors in compact output mode
|
||||
\tail <query> - live tail <query> results
|
||||
|
||||
See https://docs.victoriametrics.com/victorialogs/querying/vlogscli/ for more details
|
||||
`)
|
||||
}
|
||||
|
||||
func executeQuery(ctx context.Context, output io.Writer, qStr string, outputMode outputMode, disableColors, wrapLongLines bool) {
|
||||
func executeQuery(ctx context.Context, output io.Writer, qStr string, outputMode outputMode, wrapLongLines bool) {
|
||||
if strings.HasPrefix(qStr, `\tail `) {
|
||||
tailQuery(ctx, output, qStr, outputMode)
|
||||
return
|
||||
@@ -314,7 +293,7 @@ func executeQuery(ctx context.Context, output io.Writer, qStr string, outputMode
|
||||
_ = respBody.Close()
|
||||
}()
|
||||
|
||||
if err := readWithLess(respBody, disableColors, wrapLongLines); err != nil {
|
||||
if err := readWithLess(respBody, wrapLongLines); err != nil {
|
||||
fmt.Fprintf(output, "error when reading query response: %s\n", err)
|
||||
return
|
||||
}
|
||||
|
||||
@@ -25,8 +25,8 @@ var (
|
||||
addr = flag.String("addr", "stdout", "HTTP address to push the generated logs to; if it is set to stdout, then logs are generated to stdout")
|
||||
workers = flag.Int("workers", 1, "The number of workers to use to push logs to -addr")
|
||||
|
||||
start = newTimeFlag("start", "-1d", "Generated logs start from this time; see https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#timestamp-formats")
|
||||
end = newTimeFlag("end", "0s", "Generated logs end at this time; see https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#timestamp-formats")
|
||||
start = newTimeFlag("start", "-1d", "Generated logs start from this time; see https://docs.victoriametrics.com/#timestamp-formats")
|
||||
end = newTimeFlag("end", "0s", "Generated logs end at this time; see https://docs.victoriametrics.com/#timestamp-formats")
|
||||
activeStreams = flag.Int("activeStreams", 100, "The number of active log streams to generate; see https://docs.victoriametrics.com/victorialogs/keyconcepts/#stream-fields")
|
||||
totalStreams = flag.Int("totalStreams", 0, "The number of total log streams; if -totalStreams > -activeStreams, then some active streams are substituted with new streams "+
|
||||
"during data generation")
|
||||
@@ -206,8 +206,6 @@ func generateAndPushLogs(cfg *workerConfig, workerID int) {
|
||||
if err != nil {
|
||||
logger.Fatalf("cannot perform request to %q: %s", cfg.url, err)
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
|
||||
if resp.StatusCode/100 != 2 {
|
||||
logger.Fatalf("unexpected status code got from %q: %d; want 2xx", cfg.url, err)
|
||||
}
|
||||
|
||||
@@ -1,316 +0,0 @@
|
||||
package internalselect
|
||||
|
||||
import (
|
||||
"context"
|
||||
"fmt"
|
||||
"net/http"
|
||||
"strconv"
|
||||
"sync"
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/metrics"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage/netselect"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/atomicutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/encoding"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/encoding/zstd"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/netutil"
|
||||
)
|
||||
|
||||
// RequestHandler processes requests to /internal/select/*
|
||||
func RequestHandler(ctx context.Context, w http.ResponseWriter, r *http.Request) {
|
||||
startTime := time.Now()
|
||||
|
||||
path := r.URL.Path
|
||||
rh := requestHandlers[path]
|
||||
if rh == nil {
|
||||
httpserver.Errorf(w, r, "unsupported endpoint requested: %s", path)
|
||||
return
|
||||
}
|
||||
|
||||
metrics.GetOrCreateCounter(fmt.Sprintf(`vl_http_requests_total{path=%q}`, path)).Inc()
|
||||
if err := rh(ctx, w, r); err != nil && !netutil.IsTrivialNetworkError(err) {
|
||||
metrics.GetOrCreateCounter(fmt.Sprintf(`vl_http_request_errors_total{path=%q}`, path)).Inc()
|
||||
httpserver.Errorf(w, r, "%s", err)
|
||||
// The return is skipped intentionally in order to track the duration of failed queries.
|
||||
}
|
||||
metrics.GetOrCreateSummary(fmt.Sprintf(`vl_http_request_duration_seconds{path=%q}`, path)).UpdateDuration(startTime)
|
||||
}
|
||||
|
||||
var requestHandlers = map[string]func(ctx context.Context, w http.ResponseWriter, r *http.Request) error{
|
||||
"/internal/select/query": processQueryRequest,
|
||||
"/internal/select/field_names": processFieldNamesRequest,
|
||||
"/internal/select/field_values": processFieldValuesRequest,
|
||||
"/internal/select/stream_field_names": processStreamFieldNamesRequest,
|
||||
"/internal/select/stream_field_values": processStreamFieldValuesRequest,
|
||||
"/internal/select/streams": processStreamsRequest,
|
||||
"/internal/select/stream_ids": processStreamIDsRequest,
|
||||
}
|
||||
|
||||
func processQueryRequest(ctx context.Context, w http.ResponseWriter, r *http.Request) error {
|
||||
cp, err := getCommonParams(r, netselect.QueryProtocolVersion)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
w.Header().Set("Content-Type", "application/octet-stream")
|
||||
|
||||
var wLock sync.Mutex
|
||||
var dataLenBuf []byte
|
||||
|
||||
sendBuf := func(bb *bytesutil.ByteBuffer) error {
|
||||
if len(bb.B) == 0 {
|
||||
return nil
|
||||
}
|
||||
|
||||
data := bb.B
|
||||
if !cp.DisableCompression {
|
||||
bufLen := len(bb.B)
|
||||
bb.B = zstd.CompressLevel(bb.B, bb.B, 1)
|
||||
data = bb.B[bufLen:]
|
||||
}
|
||||
|
||||
wLock.Lock()
|
||||
dataLenBuf = encoding.MarshalUint64(dataLenBuf[:0], uint64(len(data)))
|
||||
_, err := w.Write(dataLenBuf)
|
||||
if err == nil {
|
||||
_, err = w.Write(data)
|
||||
}
|
||||
wLock.Unlock()
|
||||
|
||||
// Reset the sent buf
|
||||
bb.Reset()
|
||||
|
||||
return err
|
||||
}
|
||||
|
||||
var bufs atomicutil.Slice[bytesutil.ByteBuffer]
|
||||
|
||||
var errGlobalLock sync.Mutex
|
||||
var errGlobal error
|
||||
|
||||
writeBlock := func(workerID uint, db *logstorage.DataBlock) {
|
||||
if errGlobal != nil {
|
||||
return
|
||||
}
|
||||
|
||||
bb := bufs.Get(workerID)
|
||||
|
||||
bb.B = db.Marshal(bb.B)
|
||||
|
||||
if len(bb.B) < 1024*1024 {
|
||||
// Fast path - the bb is too small to be sent to the client yet.
|
||||
return
|
||||
}
|
||||
|
||||
// Slow path - the bb must be sent to the client.
|
||||
if err := sendBuf(bb); err != nil {
|
||||
errGlobalLock.Lock()
|
||||
if errGlobal != nil {
|
||||
errGlobal = err
|
||||
}
|
||||
errGlobalLock.Unlock()
|
||||
}
|
||||
}
|
||||
|
||||
if err := vlstorage.RunQuery(ctx, cp.TenantIDs, cp.Query, writeBlock); err != nil {
|
||||
return err
|
||||
}
|
||||
if errGlobal != nil {
|
||||
return errGlobal
|
||||
}
|
||||
|
||||
// Send the remaining data
|
||||
for _, bb := range bufs.All() {
|
||||
if err := sendBuf(bb); err != nil {
|
||||
return err
|
||||
}
|
||||
}
|
||||
|
||||
return nil
|
||||
}
|
||||
|
||||
func processFieldNamesRequest(ctx context.Context, w http.ResponseWriter, r *http.Request) error {
|
||||
cp, err := getCommonParams(r, netselect.FieldNamesProtocolVersion)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
fieldNames, err := vlstorage.GetFieldNames(ctx, cp.TenantIDs, cp.Query)
|
||||
if err != nil {
|
||||
return fmt.Errorf("cannot obtain field names: %w", err)
|
||||
}
|
||||
|
||||
return writeValuesWithHits(w, fieldNames, cp.DisableCompression)
|
||||
}
|
||||
|
||||
func processFieldValuesRequest(ctx context.Context, w http.ResponseWriter, r *http.Request) error {
|
||||
cp, err := getCommonParams(r, netselect.FieldValuesProtocolVersion)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
fieldName := r.FormValue("field")
|
||||
|
||||
limit, err := getInt64FromRequest(r, "limit")
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
fieldValues, err := vlstorage.GetFieldValues(ctx, cp.TenantIDs, cp.Query, fieldName, uint64(limit))
|
||||
if err != nil {
|
||||
return fmt.Errorf("cannot obtain field values: %w", err)
|
||||
}
|
||||
|
||||
return writeValuesWithHits(w, fieldValues, cp.DisableCompression)
|
||||
}
|
||||
|
||||
func processStreamFieldNamesRequest(ctx context.Context, w http.ResponseWriter, r *http.Request) error {
|
||||
cp, err := getCommonParams(r, netselect.StreamFieldNamesProtocolVersion)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
fieldNames, err := vlstorage.GetStreamFieldNames(ctx, cp.TenantIDs, cp.Query)
|
||||
if err != nil {
|
||||
return fmt.Errorf("cannot obtain stream field names: %w", err)
|
||||
}
|
||||
|
||||
return writeValuesWithHits(w, fieldNames, cp.DisableCompression)
|
||||
}
|
||||
|
||||
func processStreamFieldValuesRequest(ctx context.Context, w http.ResponseWriter, r *http.Request) error {
|
||||
cp, err := getCommonParams(r, netselect.StreamFieldValuesProtocolVersion)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
fieldName := r.FormValue("field")
|
||||
|
||||
limit, err := getInt64FromRequest(r, "limit")
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
fieldValues, err := vlstorage.GetStreamFieldValues(ctx, cp.TenantIDs, cp.Query, fieldName, uint64(limit))
|
||||
if err != nil {
|
||||
return fmt.Errorf("cannot obtain stream field values: %w", err)
|
||||
}
|
||||
|
||||
return writeValuesWithHits(w, fieldValues, cp.DisableCompression)
|
||||
}
|
||||
|
||||
func processStreamsRequest(ctx context.Context, w http.ResponseWriter, r *http.Request) error {
|
||||
cp, err := getCommonParams(r, netselect.StreamsProtocolVersion)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
limit, err := getInt64FromRequest(r, "limit")
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
streams, err := vlstorage.GetStreams(ctx, cp.TenantIDs, cp.Query, uint64(limit))
|
||||
if err != nil {
|
||||
return fmt.Errorf("cannot obtain streams: %w", err)
|
||||
}
|
||||
|
||||
return writeValuesWithHits(w, streams, cp.DisableCompression)
|
||||
}
|
||||
|
||||
func processStreamIDsRequest(ctx context.Context, w http.ResponseWriter, r *http.Request) error {
|
||||
cp, err := getCommonParams(r, netselect.StreamIDsProtocolVersion)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
limit, err := getInt64FromRequest(r, "limit")
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
streamIDs, err := vlstorage.GetStreamIDs(ctx, cp.TenantIDs, cp.Query, uint64(limit))
|
||||
if err != nil {
|
||||
return fmt.Errorf("cannot obtain streams: %w", err)
|
||||
}
|
||||
|
||||
return writeValuesWithHits(w, streamIDs, cp.DisableCompression)
|
||||
}
|
||||
|
||||
type commonParams struct {
|
||||
TenantIDs []logstorage.TenantID
|
||||
Query *logstorage.Query
|
||||
|
||||
DisableCompression bool
|
||||
}
|
||||
|
||||
func getCommonParams(r *http.Request, expectedProtocolVersion string) (*commonParams, error) {
|
||||
version := r.FormValue("version")
|
||||
if version != expectedProtocolVersion {
|
||||
return nil, fmt.Errorf("unexpected version=%q; want %q", version, expectedProtocolVersion)
|
||||
}
|
||||
|
||||
tenantIDsStr := r.FormValue("tenant_ids")
|
||||
tenantIDs, err := logstorage.UnmarshalTenantIDs([]byte(tenantIDsStr))
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("cannot unmarshal tenant_ids=%q: %w", tenantIDsStr, err)
|
||||
}
|
||||
|
||||
timestamp, err := getInt64FromRequest(r, "timestamp")
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
qStr := r.FormValue("query")
|
||||
q, err := logstorage.ParseQueryAtTimestamp(qStr, timestamp)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("cannot unmarshal query=%q: %w", qStr, err)
|
||||
}
|
||||
|
||||
s := r.FormValue("disable_compression")
|
||||
disableCompression, err := strconv.ParseBool(s)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("cannot parse disable_compression=%q: %w", s, err)
|
||||
}
|
||||
|
||||
cp := &commonParams{
|
||||
TenantIDs: tenantIDs,
|
||||
Query: q,
|
||||
|
||||
DisableCompression: disableCompression,
|
||||
}
|
||||
return cp, nil
|
||||
}
|
||||
|
||||
func writeValuesWithHits(w http.ResponseWriter, vhs []logstorage.ValueWithHits, disableCompression bool) error {
|
||||
var b []byte
|
||||
for i := range vhs {
|
||||
b = vhs[i].Marshal(b)
|
||||
}
|
||||
|
||||
if !disableCompression {
|
||||
b = zstd.CompressLevel(nil, b, 1)
|
||||
}
|
||||
|
||||
w.Header().Set("Content-Type", "application/octet-stream")
|
||||
|
||||
if _, err := w.Write(b); err != nil {
|
||||
return fmt.Errorf("cannot send response to the client: %w", err)
|
||||
}
|
||||
|
||||
return nil
|
||||
}
|
||||
|
||||
func getInt64FromRequest(r *http.Request, argName string) (int64, error) {
|
||||
s := r.FormValue(argName)
|
||||
n, err := strconv.ParseInt(s, 10, 64)
|
||||
if err != nil {
|
||||
return 0, fmt.Errorf("cannot parse %s=%q: %w", argName, s, err)
|
||||
}
|
||||
return n, nil
|
||||
}
|
||||
47
app/vlselect/logsql/buffered_writer.go
Normal file
47
app/vlselect/logsql/buffered_writer.go
Normal file
@@ -0,0 +1,47 @@
|
||||
package logsql
|
||||
|
||||
import (
|
||||
"bufio"
|
||||
"io"
|
||||
"sync"
|
||||
)
|
||||
|
||||
func getBufferedWriter(w io.Writer) *bufferedWriter {
|
||||
v := bufferedWriterPool.Get()
|
||||
if v == nil {
|
||||
return &bufferedWriter{
|
||||
bw: bufio.NewWriter(w),
|
||||
}
|
||||
}
|
||||
bw := v.(*bufferedWriter)
|
||||
bw.bw.Reset(w)
|
||||
return bw
|
||||
}
|
||||
|
||||
func putBufferedWriter(bw *bufferedWriter) {
|
||||
bw.reset()
|
||||
bufferedWriterPool.Put(bw)
|
||||
}
|
||||
|
||||
var bufferedWriterPool sync.Pool
|
||||
|
||||
type bufferedWriter struct {
|
||||
mu sync.Mutex
|
||||
bw *bufio.Writer
|
||||
}
|
||||
|
||||
func (bw *bufferedWriter) reset() {
|
||||
// nothing to do
|
||||
}
|
||||
|
||||
func (bw *bufferedWriter) WriteIgnoreErrors(p []byte) {
|
||||
bw.mu.Lock()
|
||||
_, _ = bw.bw.Write(p)
|
||||
bw.mu.Unlock()
|
||||
}
|
||||
|
||||
func (bw *bufferedWriter) FlushIgnoreErrors() {
|
||||
bw.mu.Lock()
|
||||
_ = bw.bw.Flush()
|
||||
bw.mu.Unlock()
|
||||
}
|
||||
@@ -3,7 +3,6 @@ package logsql
|
||||
import (
|
||||
"context"
|
||||
"fmt"
|
||||
"io"
|
||||
"math"
|
||||
"net/http"
|
||||
"regexp"
|
||||
@@ -12,14 +11,12 @@ import (
|
||||
"strconv"
|
||||
"strings"
|
||||
"sync"
|
||||
"sync/atomic"
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/metrics"
|
||||
"github.com/valyala/fastjson"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/atomicutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputil"
|
||||
@@ -55,21 +52,15 @@ func ProcessFacetsRequest(ctx context.Context, w http.ResponseWriter, r *http.Re
|
||||
}
|
||||
keepConstFields := httputil.GetBool(r, "keep_const_fields")
|
||||
|
||||
// Pipes must be dropped, since it is expected facets are obtained
|
||||
// from the real logs stored in the database.
|
||||
q.DropAllPipes()
|
||||
|
||||
q.AddFacetsPipe(limit, maxValuesPerField, maxValueLen, keepConstFields)
|
||||
|
||||
var mLock sync.Mutex
|
||||
m := make(map[string][]facetEntry)
|
||||
writeBlock := func(_ uint, db *logstorage.DataBlock) {
|
||||
rowsCount := db.RowsCount()
|
||||
if rowsCount == 0 {
|
||||
writeBlock := func(_ uint, _ []int64, columns []logstorage.BlockColumn) {
|
||||
if len(columns) == 0 || len(columns[0].Values) == 0 {
|
||||
return
|
||||
}
|
||||
|
||||
columns := db.Columns
|
||||
if len(columns) != 3 {
|
||||
logger.Panicf("BUG: expecting 3 columns; got %d columns", len(columns))
|
||||
}
|
||||
@@ -159,27 +150,23 @@ func ProcessHitsRequest(ctx context.Context, w http.ResponseWriter, r *http.Requ
|
||||
fieldsLimit = 0
|
||||
}
|
||||
|
||||
// Pipes must be dropped, since it is expected hits are obtained
|
||||
// from the real logs stored in the database.
|
||||
// Prepare the query for hits count.
|
||||
q.DropAllPipes()
|
||||
|
||||
q.AddCountByTimePipe(int64(step), int64(offset), fields)
|
||||
|
||||
var mLock sync.Mutex
|
||||
m := make(map[string]*hitsSeries)
|
||||
writeBlock := func(_ uint, db *logstorage.DataBlock) {
|
||||
rowsCount := db.RowsCount()
|
||||
if rowsCount == 0 {
|
||||
writeBlock := func(_ uint, timestamps []int64, columns []logstorage.BlockColumn) {
|
||||
if len(columns) == 0 || len(columns[0].Values) == 0 {
|
||||
return
|
||||
}
|
||||
|
||||
columns := db.Columns
|
||||
timestampValues := columns[0].Values
|
||||
hitsValues := columns[len(columns)-1].Values
|
||||
columns = columns[1 : len(columns)-1]
|
||||
|
||||
bb := blockResultPool.Get()
|
||||
for i := 0; i < rowsCount; i++ {
|
||||
for i := range timestamps {
|
||||
timestampStr := strings.Clone(timestampValues[i])
|
||||
hitsStr := strings.Clone(hitsValues[i])
|
||||
hits, err := strconv.ParseUint(hitsStr, 10, 64)
|
||||
@@ -218,8 +205,6 @@ func ProcessHitsRequest(ctx context.Context, w http.ResponseWriter, r *http.Requ
|
||||
WriteHitsSeries(w, m)
|
||||
}
|
||||
|
||||
var blockResultPool bytesutil.ByteBufferPool
|
||||
|
||||
func getTopHitsSeries(m map[string]*hitsSeries, fieldsLimit int) map[string]*hitsSeries {
|
||||
if fieldsLimit <= 0 || fieldsLimit >= len(m) {
|
||||
return m
|
||||
@@ -295,10 +280,6 @@ func ProcessFieldNamesRequest(ctx context.Context, w http.ResponseWriter, r *htt
|
||||
return
|
||||
}
|
||||
|
||||
// Pipes must be dropped, since it is expected field names are obtained
|
||||
// from the real logs stored in the database.
|
||||
q.DropAllPipes()
|
||||
|
||||
// Obtain field names for the given query
|
||||
fieldNames, err := vlstorage.GetFieldNames(ctx, tenantIDs, q)
|
||||
if err != nil {
|
||||
@@ -338,10 +319,6 @@ func ProcessFieldValuesRequest(ctx context.Context, w http.ResponseWriter, r *ht
|
||||
limit = 0
|
||||
}
|
||||
|
||||
// Pipes must be dropped, since it is expected field values are obtained
|
||||
// from the real logs stored in the database.
|
||||
q.DropAllPipes()
|
||||
|
||||
// Obtain unique values for the given field
|
||||
values, err := vlstorage.GetFieldValues(ctx, tenantIDs, q, fieldName, uint64(limit))
|
||||
if err != nil {
|
||||
@@ -364,10 +341,6 @@ func ProcessStreamFieldNamesRequest(ctx context.Context, w http.ResponseWriter,
|
||||
return
|
||||
}
|
||||
|
||||
// Pipes must be dropped, since it is expected stream field names are obtained
|
||||
// from the real logs stored in the database.
|
||||
q.DropAllPipes()
|
||||
|
||||
// Obtain stream field names for the given query
|
||||
names, err := vlstorage.GetStreamFieldNames(ctx, tenantIDs, q)
|
||||
if err != nil {
|
||||
@@ -406,10 +379,6 @@ func ProcessStreamFieldValuesRequest(ctx context.Context, w http.ResponseWriter,
|
||||
limit = 0
|
||||
}
|
||||
|
||||
// Pipes must be dropped, since it is expected stream field values are obtained
|
||||
// from the real logs stored in the database.
|
||||
q.DropAllPipes()
|
||||
|
||||
// Obtain stream field values for the given query and the given fieldName
|
||||
values, err := vlstorage.GetStreamFieldValues(ctx, tenantIDs, q, fieldName, uint64(limit))
|
||||
if err != nil {
|
||||
@@ -441,10 +410,6 @@ func ProcessStreamIDsRequest(ctx context.Context, w http.ResponseWriter, r *http
|
||||
limit = 0
|
||||
}
|
||||
|
||||
// Pipes must be dropped, since it is expected stream ids are obtained
|
||||
// from the real logs stored in the database.
|
||||
q.DropAllPipes()
|
||||
|
||||
// Obtain streamIDs for the given query
|
||||
streamIDs, err := vlstorage.GetStreamIDs(ctx, tenantIDs, q, uint64(limit))
|
||||
if err != nil {
|
||||
@@ -476,10 +441,6 @@ func ProcessStreamsRequest(ctx context.Context, w http.ResponseWriter, r *http.R
|
||||
limit = 0
|
||||
}
|
||||
|
||||
// Pipes must be dropped, since it is expected stream are obtained
|
||||
// from the real logs stored in the database.
|
||||
q.DropAllPipes()
|
||||
|
||||
// Obtain streams for the given query
|
||||
streams, err := vlstorage.GetStreams(ctx, tenantIDs, q, uint64(limit))
|
||||
if err != nil {
|
||||
@@ -543,11 +504,6 @@ func ProcessLiveTailRequest(ctx context.Context, w http.ResponseWriter, r *http.
|
||||
if !ok {
|
||||
logger.Panicf("BUG: it is expected that http.ResponseWriter (%T) supports http.Flusher interface", w)
|
||||
}
|
||||
|
||||
w.Header().Set("Content-Type", "application/x-ndjson")
|
||||
w.Header().Set("Access-Control-Allow-Origin", "*")
|
||||
flusher.Flush()
|
||||
|
||||
qOrig := q
|
||||
for {
|
||||
q = qOrig.CloneWithTimeFilter(end, start, end)
|
||||
@@ -610,8 +566,8 @@ func newTailProcessor(cancel func()) *tailProcessor {
|
||||
}
|
||||
}
|
||||
|
||||
func (tp *tailProcessor) writeBlock(_ uint, db *logstorage.DataBlock) {
|
||||
if db.RowsCount() == 0 {
|
||||
func (tp *tailProcessor) writeBlock(_ uint, timestamps []int64, columns []logstorage.BlockColumn) {
|
||||
if len(timestamps) == 0 {
|
||||
return
|
||||
}
|
||||
|
||||
@@ -623,8 +579,14 @@ func (tp *tailProcessor) writeBlock(_ uint, db *logstorage.DataBlock) {
|
||||
}
|
||||
|
||||
// Make sure columns contain _time field, since it is needed for proper tail work.
|
||||
timestamps, ok := db.GetTimestamps(nil)
|
||||
if !ok {
|
||||
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
|
||||
@@ -633,8 +595,8 @@ func (tp *tailProcessor) writeBlock(_ uint, db *logstorage.DataBlock) {
|
||||
// Copy block rows to tp.perStreamRows
|
||||
for i, timestamp := range timestamps {
|
||||
streamID := ""
|
||||
fields := make([]logstorage.Field, len(db.Columns))
|
||||
for j, c := range db.Columns {
|
||||
fields := make([]logstorage.Field, len(columns))
|
||||
for j, c := range columns {
|
||||
name := strings.Clone(c.Name)
|
||||
value := strings.Clone(c.Values[i])
|
||||
|
||||
@@ -726,15 +688,12 @@ func ProcessStatsQueryRangeRequest(ctx context.Context, w http.ResponseWriter, r
|
||||
m := make(map[string]*statsSeries)
|
||||
var mLock sync.Mutex
|
||||
|
||||
writeBlock := func(_ uint, db *logstorage.DataBlock) {
|
||||
rowsCount := db.RowsCount()
|
||||
|
||||
columns := db.Columns
|
||||
writeBlock := func(_ uint, timestamps []int64, columns []logstorage.BlockColumn) {
|
||||
clonedColumnNames := make([]string, len(columns))
|
||||
for i, c := range columns {
|
||||
clonedColumnNames[i] = strings.Clone(c.Name)
|
||||
}
|
||||
for i := 0; i < rowsCount; i++ {
|
||||
for i := range timestamps {
|
||||
// Do not move q.GetTimestamp() outside writeBlock, since ts
|
||||
// must be initialized to query timestamp for every processed log row.
|
||||
// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/8312
|
||||
@@ -843,14 +802,12 @@ func ProcessStatsQueryRequest(ctx context.Context, w http.ResponseWriter, r *htt
|
||||
var rowsLock sync.Mutex
|
||||
|
||||
timestamp := q.GetTimestamp()
|
||||
writeBlock := func(_ uint, db *logstorage.DataBlock) {
|
||||
rowsCount := db.RowsCount()
|
||||
columns := db.Columns
|
||||
writeBlock := func(_ uint, timestamps []int64, columns []logstorage.BlockColumn) {
|
||||
clonedColumnNames := make([]string, len(columns))
|
||||
for i, c := range columns {
|
||||
clonedColumnNames[i] = strings.Clone(c.Name)
|
||||
}
|
||||
for i := 0; i < rowsCount; i++ {
|
||||
for i := range timestamps {
|
||||
labels := make([]logstorage.Field, 0, len(byFields))
|
||||
for j, c := range columns {
|
||||
if slices.Contains(byFields, c.Name) {
|
||||
@@ -912,21 +869,11 @@ func ProcessQueryRequest(ctx context.Context, w http.ResponseWriter, r *http.Req
|
||||
return
|
||||
}
|
||||
|
||||
sw := &syncWriter{
|
||||
w: w,
|
||||
}
|
||||
|
||||
var bwShards atomicutil.Slice[bufferedWriter]
|
||||
bwShards.Init = func(shard *bufferedWriter) {
|
||||
shard.sw = sw
|
||||
}
|
||||
bw := getBufferedWriter(w)
|
||||
defer func() {
|
||||
shards := bwShards.All()
|
||||
for _, shard := range shards {
|
||||
shard.FlushIgnoreErrors()
|
||||
}
|
||||
bw.FlushIgnoreErrors()
|
||||
putBufferedWriter(bw)
|
||||
}()
|
||||
|
||||
w.Header().Set("Content-Type", "application/stream+json")
|
||||
|
||||
if limit > 0 {
|
||||
@@ -936,34 +883,32 @@ func ProcessQueryRequest(ctx context.Context, w http.ResponseWriter, r *http.Req
|
||||
httpserver.Errorf(w, r, "%s", err)
|
||||
return
|
||||
}
|
||||
bw := bwShards.Get(0)
|
||||
bb := blockResultPool.Get()
|
||||
b := bb.B
|
||||
for i := range rows {
|
||||
bw.buf = logstorage.MarshalFieldsToJSON(bw.buf, rows[i].fields)
|
||||
bw.buf = append(bw.buf, '\n')
|
||||
if len(bw.buf) > 16*1024 {
|
||||
bw.FlushIgnoreErrors()
|
||||
}
|
||||
b = logstorage.MarshalFieldsToJSON(b[:0], rows[i].fields)
|
||||
b = append(b, '\n')
|
||||
bw.WriteIgnoreErrors(b)
|
||||
}
|
||||
bb.B = b
|
||||
blockResultPool.Put(bb)
|
||||
return
|
||||
}
|
||||
|
||||
q.AddPipeLimit(uint64(limit))
|
||||
}
|
||||
|
||||
writeBlock := func(workerID uint, db *logstorage.DataBlock) {
|
||||
rowsCount := db.RowsCount()
|
||||
if rowsCount == 0 {
|
||||
writeBlock := func(_ uint, timestamps []int64, columns []logstorage.BlockColumn) {
|
||||
if len(columns) == 0 || len(columns[0].Values) == 0 {
|
||||
return
|
||||
}
|
||||
columns := db.Columns
|
||||
|
||||
bw := bwShards.Get(workerID)
|
||||
for i := 0; i < rowsCount; i++ {
|
||||
WriteJSONRow(bw, columns, i)
|
||||
if len(bw.buf) > 16*1024 {
|
||||
bw.FlushIgnoreErrors()
|
||||
}
|
||||
bb := blockResultPool.Get()
|
||||
for i := range timestamps {
|
||||
WriteJSONRow(bb, columns, i)
|
||||
}
|
||||
bw.WriteIgnoreErrors(bb.B)
|
||||
blockResultPool.Put(bb)
|
||||
}
|
||||
|
||||
if err := vlstorage.RunQuery(ctx, tenantIDs, q, writeBlock); err != nil {
|
||||
@@ -972,37 +917,14 @@ func ProcessQueryRequest(ctx context.Context, w http.ResponseWriter, r *http.Req
|
||||
}
|
||||
}
|
||||
|
||||
type syncWriter struct {
|
||||
mu sync.Mutex
|
||||
w io.Writer
|
||||
var blockResultPool bytesutil.ByteBufferPool
|
||||
|
||||
type row struct {
|
||||
timestamp int64
|
||||
fields []logstorage.Field
|
||||
}
|
||||
|
||||
func (sw *syncWriter) Write(p []byte) (int, error) {
|
||||
sw.mu.Lock()
|
||||
n, err := sw.w.Write(p)
|
||||
sw.mu.Unlock()
|
||||
return n, err
|
||||
}
|
||||
|
||||
type bufferedWriter struct {
|
||||
buf []byte
|
||||
sw *syncWriter
|
||||
}
|
||||
|
||||
func (bw *bufferedWriter) Write(p []byte) (int, error) {
|
||||
bw.buf = append(bw.buf, p...)
|
||||
|
||||
// Do not send bw.buf to bw.sw here, since the data at bw.buf may be incomplete (it must end with '\n')
|
||||
|
||||
return len(p), nil
|
||||
}
|
||||
|
||||
func (bw *bufferedWriter) FlushIgnoreErrors() {
|
||||
_, _ = bw.sw.Write(bw.buf)
|
||||
bw.buf = bw.buf[:0]
|
||||
}
|
||||
|
||||
func getLastNQueryResults(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, limit int) ([]logRow, error) {
|
||||
func getLastNQueryResults(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, limit int) ([]row, error) {
|
||||
limitUpper := 2 * limit
|
||||
q.AddPipeLimit(uint64(limitUpper))
|
||||
|
||||
@@ -1071,39 +993,28 @@ func getLastNQueryResults(ctx context.Context, tenantIDs []logstorage.TenantID,
|
||||
}
|
||||
}
|
||||
|
||||
func getLastNRows(rows []logRow, limit int) []logRow {
|
||||
sortLogRows(rows)
|
||||
func getLastNRows(rows []row, limit int) []row {
|
||||
sort.Slice(rows, func(i, j int) bool {
|
||||
return rows[i].timestamp < rows[j].timestamp
|
||||
})
|
||||
if len(rows) > limit {
|
||||
rows = rows[len(rows)-limit:]
|
||||
}
|
||||
return rows
|
||||
}
|
||||
|
||||
func getQueryResultsWithLimit(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, limit int) ([]logRow, error) {
|
||||
func getQueryResultsWithLimit(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, limit int) ([]row, error) {
|
||||
ctxWithCancel, cancel := context.WithCancel(ctx)
|
||||
defer cancel()
|
||||
|
||||
var missingTimeColumn atomic.Bool
|
||||
var rows []logRow
|
||||
var rows []row
|
||||
var rowsLock sync.Mutex
|
||||
writeBlock := func(_ uint, db *logstorage.DataBlock) {
|
||||
if missingTimeColumn.Load() {
|
||||
return
|
||||
}
|
||||
|
||||
columns := db.Columns
|
||||
writeBlock := func(_ uint, timestamps []int64, columns []logstorage.BlockColumn) {
|
||||
clonedColumnNames := make([]string, len(columns))
|
||||
for i, c := range columns {
|
||||
clonedColumnNames[i] = strings.Clone(c.Name)
|
||||
}
|
||||
|
||||
timestamps, ok := db.GetTimestamps(nil)
|
||||
if !ok {
|
||||
missingTimeColumn.Store(true)
|
||||
cancel()
|
||||
return
|
||||
}
|
||||
|
||||
for i, timestamp := range timestamps {
|
||||
fields := make([]logstorage.Field, len(columns))
|
||||
for j := range columns {
|
||||
@@ -1113,7 +1024,7 @@ func getQueryResultsWithLimit(ctx context.Context, tenantIDs []logstorage.Tenant
|
||||
}
|
||||
|
||||
rowsLock.Lock()
|
||||
rows = append(rows, logRow{
|
||||
rows = append(rows, row{
|
||||
timestamp: timestamp,
|
||||
fields: fields,
|
||||
})
|
||||
@@ -1124,13 +1035,11 @@ func getQueryResultsWithLimit(ctx context.Context, tenantIDs []logstorage.Tenant
|
||||
cancel()
|
||||
}
|
||||
}
|
||||
err := vlstorage.RunQuery(ctxWithCancel, tenantIDs, q, writeBlock)
|
||||
|
||||
if missingTimeColumn.Load() {
|
||||
return nil, fmt.Errorf("missing _time column in the result for the query [%s]", q)
|
||||
if err := vlstorage.RunQuery(ctxWithCancel, tenantIDs, q, writeBlock); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
return rows, err
|
||||
return rows, nil
|
||||
}
|
||||
|
||||
func parseCommonArgs(r *http.Request) (*logstorage.Query, []logstorage.TenantID, error) {
|
||||
@@ -1189,22 +1098,20 @@ func parseCommonArgs(r *http.Request) (*logstorage.Query, []logstorage.TenantID,
|
||||
}
|
||||
|
||||
// Parse optional extra_filters
|
||||
for _, extraFiltersStr := range r.Form["extra_filters"] {
|
||||
extraFilters, err := parseExtraFilters(extraFiltersStr)
|
||||
if err != nil {
|
||||
return nil, nil, err
|
||||
}
|
||||
q.AddExtraFilters(extraFilters)
|
||||
extraFiltersStr := r.FormValue("extra_filters")
|
||||
extraFilters, err := parseExtraFilters(extraFiltersStr)
|
||||
if err != nil {
|
||||
return nil, nil, err
|
||||
}
|
||||
q.AddExtraFilters(extraFilters)
|
||||
|
||||
// Parse optional extra_stream_filters
|
||||
for _, extraStreamFiltersStr := range r.Form["extra_stream_filters"] {
|
||||
extraStreamFilters, err := parseExtraStreamFilters(extraStreamFiltersStr)
|
||||
if err != nil {
|
||||
return nil, nil, err
|
||||
}
|
||||
q.AddExtraFilters(extraStreamFilters)
|
||||
extraStreamFiltersStr := r.FormValue("extra_stream_filters")
|
||||
extraStreamFilters, err := parseExtraStreamFilters(extraStreamFiltersStr)
|
||||
if err != nil {
|
||||
return nil, nil, err
|
||||
}
|
||||
q.AddExtraFilters(extraStreamFilters)
|
||||
|
||||
return q, tenantIDs, nil
|
||||
}
|
||||
|
||||
@@ -9,7 +9,6 @@ import (
|
||||
"strings"
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlselect/internalselect"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlselect/logsql"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/cgroup"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
@@ -25,9 +24,6 @@ var (
|
||||
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 to a smaller value on a per-query basis via 'timeout' query arg")
|
||||
|
||||
disableSelect = flag.Bool("select.disable", false, "Whether to disable /select/* HTTP endpoints")
|
||||
disableInternal = flag.Bool("internalselect.disable", false, "Whether to disable /internal/select/* HTTP endpoints")
|
||||
)
|
||||
|
||||
func getDefaultMaxConcurrentRequests() int {
|
||||
@@ -74,31 +70,13 @@ var vmuiFileServer = http.FileServer(http.FS(vmuiFiles))
|
||||
|
||||
// RequestHandler handles select requests for VictoriaLogs
|
||||
func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
|
||||
path := strings.ReplaceAll(r.URL.Path, "//", "/")
|
||||
path := r.URL.Path
|
||||
|
||||
if strings.HasPrefix(path, "/select/") {
|
||||
if *disableSelect {
|
||||
httpserver.Errorf(w, r, "requests to /select/* are disabled with -select.disable command-line flag")
|
||||
return true
|
||||
}
|
||||
|
||||
return selectHandler(w, r, path)
|
||||
if !strings.HasPrefix(path, "/select/") {
|
||||
// Skip requests, which do not start with /select/, since these aren't our requests.
|
||||
return false
|
||||
}
|
||||
|
||||
if strings.HasPrefix(path, "/internal/select/") {
|
||||
if *disableInternal || *disableSelect {
|
||||
httpserver.Errorf(w, r, "requests to /internal/select/* are disabled with -internalselect.disable or -select.disable command-line flag")
|
||||
return true
|
||||
}
|
||||
internalselect.RequestHandler(r.Context(), w, r)
|
||||
return true
|
||||
}
|
||||
|
||||
return false
|
||||
}
|
||||
|
||||
func selectHandler(w http.ResponseWriter, r *http.Request, path string) bool {
|
||||
ctx := r.Context()
|
||||
path = strings.ReplaceAll(path, "//", "/")
|
||||
|
||||
if path == "/select/vmui" {
|
||||
// VMUI access via incomplete url without `/` in the end. Redirect to complete url.
|
||||
@@ -121,16 +99,9 @@ func selectHandler(w http.ResponseWriter, r *http.Request, path string) bool {
|
||||
return true
|
||||
}
|
||||
|
||||
if path == "/select/logsql/tail" {
|
||||
logsqlTailRequests.Inc()
|
||||
// Process live tailing request without timeout, since it is OK to run live tailing requests for very long time.
|
||||
// Also do not apply concurrency limit to tail requests, since these limits are intended for non-tail requests.
|
||||
logsql.ProcessLiveTailRequest(ctx, w, r)
|
||||
return true
|
||||
}
|
||||
|
||||
// Limit the number of concurrent queries, which can consume big amounts of CPU time.
|
||||
startTime := time.Now()
|
||||
ctx := r.Context()
|
||||
d := getMaxQueryDuration(r)
|
||||
ctxWithTimeout, cancel := context.WithTimeout(ctx, d)
|
||||
defer cancel()
|
||||
@@ -140,22 +111,28 @@ func selectHandler(w http.ResponseWriter, r *http.Request, path string) bool {
|
||||
}
|
||||
defer decRequestConcurrency()
|
||||
|
||||
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
|
||||
}
|
||||
|
||||
logRequestErrorIfNeeded(ctxWithTimeout, w, r, startTime)
|
||||
return true
|
||||
}
|
||||
|
||||
func logRequestErrorIfNeeded(ctx context.Context, w http.ResponseWriter, r *http.Request, startTime time.Time) {
|
||||
err := ctx.Err()
|
||||
err := ctxWithTimeout.Err()
|
||||
switch err {
|
||||
case nil:
|
||||
// nothing to do
|
||||
case context.Canceled:
|
||||
// do not log canceled requests, since they are expected and legal.
|
||||
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: "+
|
||||
@@ -166,6 +143,8 @@ func logRequestErrorIfNeeded(ctx context.Context, w http.ResponseWriter, r *http
|
||||
default:
|
||||
httpserver.Errorf(w, r, "unexpected error: %s", err)
|
||||
}
|
||||
|
||||
return true
|
||||
}
|
||||
|
||||
func incRequestConcurrency(ctx context.Context, w http.ResponseWriter, r *http.Request) bool {
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
File diff suppressed because one or more lines are too long
204
app/vlselect/vmui/assets/index-BgdvCSTM.js
Normal file
204
app/vlselect/vmui/assets/index-BgdvCSTM.js
Normal file
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
1
app/vlselect/vmui/assets/index-u4IOGr0E.css
Normal file
1
app/vlselect/vmui/assets/index-u4IOGr0E.css
Normal file
File diff suppressed because one or more lines are too long
76
app/vlselect/vmui/assets/vendor-DojlIpLz.js
Normal file
76
app/vlselect/vmui/assets/vendor-DojlIpLz.js
Normal file
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
@@ -35,10 +35,10 @@
|
||||
<meta property="og:title" content="UI for VictoriaLogs">
|
||||
<meta property="og:url" content="https://victoriametrics.com/products/victorialogs/">
|
||||
<meta property="og:description" content="Explore your log data with VictoriaLogs UI">
|
||||
<script type="module" crossorigin src="./assets/index-721xTF8u.js"></script>
|
||||
<link rel="modulepreload" crossorigin href="./assets/vendor-V4vnRsM-.js">
|
||||
<script type="module" crossorigin src="./assets/index-BgdvCSTM.js"></script>
|
||||
<link rel="modulepreload" crossorigin href="./assets/vendor-DojlIpLz.js">
|
||||
<link rel="stylesheet" crossorigin href="./assets/vendor-D1GxaB_c.css">
|
||||
<link rel="stylesheet" crossorigin href="./assets/index-C36SC0pJ.css">
|
||||
<link rel="stylesheet" crossorigin href="./assets/index-u4IOGr0E.css">
|
||||
</head>
|
||||
<body>
|
||||
<noscript>You need to enable JavaScript to run this app.</noscript>
|
||||
|
||||
@@ -10,14 +10,11 @@ import (
|
||||
|
||||
"github.com/VictoriaMetrics/metrics"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage/netinsert"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage/netselect"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/fs"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/promauth"
|
||||
)
|
||||
|
||||
var (
|
||||
@@ -41,59 +38,15 @@ var (
|
||||
minFreeDiskSpaceBytes = flagutil.NewBytes("storage.minFreeDiskSpaceBytes", 10e6, "The minimum free disk space at -storageDataPath after which "+
|
||||
"the storage stops accepting new data")
|
||||
|
||||
forceMergeAuthKey = flagutil.NewPassword("forceMergeAuthKey", "authKey, which must be passed in query string to /internal/force_merge . It overrides -httpAuth.* . "+
|
||||
"See https://docs.victoriametrics.com/victorialogs/#forced-merge")
|
||||
forceFlushAuthKey = flagutil.NewPassword("forceFlushAuthKey", "authKey, which must be passed in query string to /internal/force_flush . It overrides -httpAuth.* . "+
|
||||
"See https://docs.victoriametrics.com/victorialogs/#forced-flush")
|
||||
|
||||
storageNodeAddrs = flagutil.NewArrayString("storageNode", "Comma-separated list of TCP addresses for storage nodes to route the ingested logs to and to send select queries to. "+
|
||||
"If the list is empty, then the ingested logs are stored and queried locally from -storageDataPath")
|
||||
insertConcurrency = flag.Int("insert.concurrency", 2, "The average number of concurrent data ingestion requests, which can be sent to every -storageNode")
|
||||
insertDisableCompression = flag.Bool("insert.disableCompression", false, "Whether to disable compression when sending the ingested data to -storageNode nodes. "+
|
||||
"Disabled compression reduces CPU usage at the cost of higher network usage")
|
||||
selectDisableCompression = flag.Bool("select.disableCompression", false, "Whether to disable compression for select query responses received from -storageNode nodes. "+
|
||||
"Disabled compression reduces CPU usage at the cost of higher network usage")
|
||||
|
||||
storageNodeUsername = flagutil.NewArrayString("storageNode.username", "Optional basic auth username to use for the corresponding -storageNode")
|
||||
storageNodePassword = flagutil.NewArrayString("storageNode.password", "Optional basic auth password to use for the corresponding -storageNode")
|
||||
storageNodePasswordFile = flagutil.NewArrayString("storageNode.passwordFile", "Optional path to basic auth password to use for the corresponding -storageNode. "+
|
||||
"The file is re-read every second")
|
||||
storageNodeBearerToken = flagutil.NewArrayString("storageNode.bearerToken", "Optional bearer auth token to use for the corresponding -storageNode")
|
||||
storageNodeBearerTokenFile = flagutil.NewArrayString("storageNode.bearerTokenFile", "Optional path to bearer token file to use for the corresponding -storageNode. "+
|
||||
"The token is re-read from the file every second")
|
||||
|
||||
storageNodeTLS = flagutil.NewArrayBool("storageNode.tls", "Whether to use TLS (HTTPS) protocol for communicating with the corresponding -storageNode. "+
|
||||
"By default communication is performed via HTTP")
|
||||
storageNodeTLSCAFile = flagutil.NewArrayString("storageNode.tlsCAFile", "Optional path to TLS CA file to use for verifying connections to the corresponding -storageNode. "+
|
||||
"By default, system CA is used")
|
||||
storageNodeTLSCertFile = flagutil.NewArrayString("storageNode.tlsCertFile", "Optional path to client-side TLS certificate file to use when connecting "+
|
||||
"to the corresponding -storageNode")
|
||||
storageNodeTLSKeyFile = flagutil.NewArrayString("storageNode.tlsKeyFile", "Optional path to client-side TLS certificate key to use when connecting to the corresponding -storageNode")
|
||||
storageNodeTLSServerName = flagutil.NewArrayString("storageNode.tlsServerName", "Optional TLS server name to use for connections to the corresponding -storageNode. "+
|
||||
"By default, the server name from -storageNode is used")
|
||||
storageNodeTLSInsecureSkipVerify = flagutil.NewArrayBool("storageNode.tlsInsecureSkipVerify", "Whether to skip tls verification when connecting to the corresponding -storageNode")
|
||||
forceMergeAuthKey = flagutil.NewPassword("forceMergeAuthKey", "authKey, which must be passed in query string to /internal/force_merge pages. It overrides -httpAuth.*")
|
||||
)
|
||||
|
||||
var localStorage *logstorage.Storage
|
||||
var localStorageMetrics *metrics.Set
|
||||
|
||||
var netstorageInsert *netinsert.Storage
|
||||
var netstorageSelect *netselect.Storage
|
||||
|
||||
// Init initializes vlstorage.
|
||||
//
|
||||
// Stop must be called when vlstorage is no longer needed
|
||||
func Init() {
|
||||
if len(*storageNodeAddrs) == 0 {
|
||||
initLocalStorage()
|
||||
} else {
|
||||
initNetworkStorage()
|
||||
}
|
||||
}
|
||||
|
||||
func initLocalStorage() {
|
||||
if localStorage != nil {
|
||||
logger.Panicf("BUG: initLocalStorage() has been already called")
|
||||
if strg != nil {
|
||||
logger.Panicf("BUG: Init() has been already called")
|
||||
}
|
||||
|
||||
if retentionPeriod.Duration() < 24*time.Hour {
|
||||
@@ -110,160 +63,60 @@ func initLocalStorage() {
|
||||
}
|
||||
logger.Infof("opening storage at -storageDataPath=%s", *storageDataPath)
|
||||
startTime := time.Now()
|
||||
localStorage = logstorage.MustOpenStorage(*storageDataPath, cfg)
|
||||
strg = logstorage.MustOpenStorage(*storageDataPath, cfg)
|
||||
|
||||
var ss logstorage.StorageStats
|
||||
localStorage.UpdateStats(&ss)
|
||||
strg.UpdateStats(&ss)
|
||||
logger.Infof("successfully opened storage in %.3f seconds; smallParts: %d; bigParts: %d; smallPartBlocks: %d; bigPartBlocks: %d; smallPartRows: %d; bigPartRows: %d; "+
|
||||
"smallPartSize: %d bytes; bigPartSize: %d bytes",
|
||||
time.Since(startTime).Seconds(), ss.SmallParts, ss.BigParts, ss.SmallPartBlocks, ss.BigPartBlocks, ss.SmallPartRowsCount, ss.BigPartRowsCount,
|
||||
ss.CompressedSmallPartSize, ss.CompressedBigPartSize)
|
||||
|
||||
// register local storage metrics
|
||||
localStorageMetrics = metrics.NewSet()
|
||||
localStorageMetrics.RegisterMetricsWriter(func(w io.Writer) {
|
||||
writeStorageMetrics(w, localStorage)
|
||||
// register storage metrics
|
||||
storageMetrics = metrics.NewSet()
|
||||
storageMetrics.RegisterMetricsWriter(func(w io.Writer) {
|
||||
writeStorageMetrics(w, strg)
|
||||
})
|
||||
metrics.RegisterSet(localStorageMetrics)
|
||||
}
|
||||
|
||||
func initNetworkStorage() {
|
||||
if netstorageInsert != nil || netstorageSelect != nil {
|
||||
logger.Panicf("BUG: initNetworkStorage() has been already called")
|
||||
}
|
||||
|
||||
authCfgs := make([]*promauth.Config, len(*storageNodeAddrs))
|
||||
isTLSs := make([]bool, len(*storageNodeAddrs))
|
||||
for i := range authCfgs {
|
||||
authCfgs[i] = newAuthConfigForStorageNode(i)
|
||||
isTLSs[i] = storageNodeTLS.GetOptionalArg(i)
|
||||
}
|
||||
|
||||
logger.Infof("starting insert service for nodes %s", *storageNodeAddrs)
|
||||
netstorageInsert = netinsert.NewStorage(*storageNodeAddrs, authCfgs, isTLSs, *insertConcurrency, *insertDisableCompression)
|
||||
|
||||
logger.Infof("initializing select service for nodes %s", *storageNodeAddrs)
|
||||
netstorageSelect = netselect.NewStorage(*storageNodeAddrs, authCfgs, isTLSs, *selectDisableCompression)
|
||||
|
||||
logger.Infof("initialized all the network services")
|
||||
}
|
||||
|
||||
func newAuthConfigForStorageNode(argIdx int) *promauth.Config {
|
||||
username := storageNodeUsername.GetOptionalArg(argIdx)
|
||||
password := storageNodePassword.GetOptionalArg(argIdx)
|
||||
passwordFile := storageNodePasswordFile.GetOptionalArg(argIdx)
|
||||
var basicAuthCfg *promauth.BasicAuthConfig
|
||||
if username != "" || password != "" || passwordFile != "" {
|
||||
basicAuthCfg = &promauth.BasicAuthConfig{
|
||||
Username: username,
|
||||
Password: promauth.NewSecret(password),
|
||||
PasswordFile: passwordFile,
|
||||
}
|
||||
}
|
||||
|
||||
token := storageNodeBearerToken.GetOptionalArg(argIdx)
|
||||
tokenFile := storageNodeBearerTokenFile.GetOptionalArg(argIdx)
|
||||
|
||||
tlsCfg := &promauth.TLSConfig{
|
||||
CAFile: storageNodeTLSCAFile.GetOptionalArg(argIdx),
|
||||
CertFile: storageNodeTLSCertFile.GetOptionalArg(argIdx),
|
||||
KeyFile: storageNodeTLSKeyFile.GetOptionalArg(argIdx),
|
||||
ServerName: storageNodeTLSServerName.GetOptionalArg(argIdx),
|
||||
InsecureSkipVerify: storageNodeTLSInsecureSkipVerify.GetOptionalArg(argIdx),
|
||||
}
|
||||
|
||||
opts := &promauth.Options{
|
||||
BasicAuth: basicAuthCfg,
|
||||
BearerToken: token,
|
||||
BearerTokenFile: tokenFile,
|
||||
TLSConfig: tlsCfg,
|
||||
}
|
||||
ac, err := opts.NewConfig()
|
||||
if err != nil {
|
||||
logger.Panicf("FATAL: cannot populate auth config for storage node #%d: %s", argIdx, err)
|
||||
}
|
||||
|
||||
return ac
|
||||
metrics.RegisterSet(storageMetrics)
|
||||
}
|
||||
|
||||
// Stop stops vlstorage.
|
||||
func Stop() {
|
||||
if localStorage != nil {
|
||||
metrics.UnregisterSet(localStorageMetrics, true)
|
||||
localStorageMetrics = nil
|
||||
metrics.UnregisterSet(storageMetrics, true)
|
||||
storageMetrics = nil
|
||||
|
||||
localStorage.MustClose()
|
||||
localStorage = nil
|
||||
} else {
|
||||
netstorageInsert.MustStop()
|
||||
netstorageInsert = nil
|
||||
|
||||
netstorageSelect.MustStop()
|
||||
netstorageSelect = nil
|
||||
}
|
||||
strg.MustClose()
|
||||
strg = nil
|
||||
}
|
||||
|
||||
// RequestHandler is a storage request handler.
|
||||
func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
|
||||
path := r.URL.Path
|
||||
switch path {
|
||||
case "/internal/force_merge":
|
||||
return processForceMerge(w, r)
|
||||
case "/internal/force_flush":
|
||||
return processForceFlush(w, r)
|
||||
if path == "/internal/force_merge" {
|
||||
if !httpserver.CheckAuthFlag(w, r, forceMergeAuthKey) {
|
||||
return true
|
||||
}
|
||||
// Run force merge in background
|
||||
partitionNamePrefix := r.FormValue("partition_prefix")
|
||||
go func() {
|
||||
activeForceMerges.Inc()
|
||||
defer activeForceMerges.Dec()
|
||||
logger.Infof("forced merge for partition_prefix=%q has been started", partitionNamePrefix)
|
||||
startTime := time.Now()
|
||||
strg.MustForceMerge(partitionNamePrefix)
|
||||
logger.Infof("forced merge for partition_prefix=%q has been successfully finished in %.3f seconds", partitionNamePrefix, time.Since(startTime).Seconds())
|
||||
}()
|
||||
return true
|
||||
}
|
||||
return false
|
||||
}
|
||||
|
||||
func processForceMerge(w http.ResponseWriter, r *http.Request) bool {
|
||||
if localStorage == nil {
|
||||
// Force merge isn't supported by non-local storage
|
||||
return false
|
||||
}
|
||||
var strg *logstorage.Storage
|
||||
var storageMetrics *metrics.Set
|
||||
|
||||
if !httpserver.CheckAuthFlag(w, r, forceMergeAuthKey) {
|
||||
return true
|
||||
}
|
||||
|
||||
// Run force merge in background
|
||||
partitionNamePrefix := r.FormValue("partition_prefix")
|
||||
go func() {
|
||||
activeForceMerges.Inc()
|
||||
defer activeForceMerges.Dec()
|
||||
logger.Infof("forced merge for partition_prefix=%q has been started", partitionNamePrefix)
|
||||
startTime := time.Now()
|
||||
localStorage.MustForceMerge(partitionNamePrefix)
|
||||
logger.Infof("forced merge for partition_prefix=%q has been successfully finished in %.3f seconds", partitionNamePrefix, time.Since(startTime).Seconds())
|
||||
}()
|
||||
return true
|
||||
}
|
||||
|
||||
func processForceFlush(w http.ResponseWriter, r *http.Request) bool {
|
||||
if localStorage == nil {
|
||||
// Force merge isn't supported by non-local storage
|
||||
return false
|
||||
}
|
||||
|
||||
if !httpserver.CheckAuthFlag(w, r, forceFlushAuthKey) {
|
||||
return true
|
||||
}
|
||||
|
||||
logger.Infof("flushing storage to make pending data available for reading")
|
||||
localStorage.DebugFlush()
|
||||
return true
|
||||
}
|
||||
|
||||
// Storage implements insertutil.LogRowsStorage interface
|
||||
type Storage struct{}
|
||||
|
||||
// CanWriteData returns non-nil error if it cannot write data to vlstorage
|
||||
func (*Storage) CanWriteData() error {
|
||||
if localStorage == nil {
|
||||
// The data can be always written in non-local mode.
|
||||
return nil
|
||||
}
|
||||
|
||||
if localStorage.IsReadOnly() {
|
||||
// CanWriteData returns non-nil error if it cannot write data to vlstorage.
|
||||
func CanWriteData() error {
|
||||
if strg.IsReadOnly() {
|
||||
return &httpserver.ErrorWithStatusCode{
|
||||
Err: fmt.Errorf("cannot add rows into storage in read-only mode; the storage can be in read-only mode "+
|
||||
"because of lack of free disk space at -storageDataPath=%s", *storageDataPath),
|
||||
@@ -276,78 +129,51 @@ func (*Storage) CanWriteData() error {
|
||||
// MustAddRows adds lr to vlstorage
|
||||
//
|
||||
// It is advised to call CanWriteData() before calling MustAddRows()
|
||||
func (*Storage) MustAddRows(lr *logstorage.LogRows) {
|
||||
if localStorage != nil {
|
||||
// Store lr in the local storage.
|
||||
localStorage.MustAddRows(lr)
|
||||
} else {
|
||||
// Store lr across the remote storage nodes.
|
||||
lr.ForEachRow(netstorageInsert.AddRow)
|
||||
}
|
||||
func MustAddRows(lr *logstorage.LogRows) {
|
||||
strg.MustAddRows(lr)
|
||||
}
|
||||
|
||||
// RunQuery runs the given q and calls writeBlock for the returned data blocks
|
||||
func RunQuery(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, writeBlock logstorage.WriteDataBlockFunc) error {
|
||||
if localStorage != nil {
|
||||
return localStorage.RunQuery(ctx, tenantIDs, q, writeBlock)
|
||||
}
|
||||
return netstorageSelect.RunQuery(ctx, tenantIDs, q, writeBlock)
|
||||
func RunQuery(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, writeBlock logstorage.WriteBlockFunc) error {
|
||||
return strg.RunQuery(ctx, tenantIDs, q, writeBlock)
|
||||
}
|
||||
|
||||
// GetFieldNames executes q and returns field names seen in results.
|
||||
func GetFieldNames(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query) ([]logstorage.ValueWithHits, error) {
|
||||
if localStorage != nil {
|
||||
return localStorage.GetFieldNames(ctx, tenantIDs, q)
|
||||
}
|
||||
return netstorageSelect.GetFieldNames(ctx, tenantIDs, q)
|
||||
return strg.GetFieldNames(ctx, tenantIDs, q)
|
||||
}
|
||||
|
||||
// GetFieldValues executes q and returns unique values for the fieldName seen in results.
|
||||
//
|
||||
// If limit > 0, then up to limit unique values are returned.
|
||||
func GetFieldValues(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, fieldName string, limit uint64) ([]logstorage.ValueWithHits, error) {
|
||||
if localStorage != nil {
|
||||
return localStorage.GetFieldValues(ctx, tenantIDs, q, fieldName, limit)
|
||||
}
|
||||
return netstorageSelect.GetFieldValues(ctx, tenantIDs, q, fieldName, limit)
|
||||
return strg.GetFieldValues(ctx, tenantIDs, q, fieldName, limit)
|
||||
}
|
||||
|
||||
// GetStreamFieldNames executes q and returns stream field names seen in results.
|
||||
func GetStreamFieldNames(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query) ([]logstorage.ValueWithHits, error) {
|
||||
if localStorage != nil {
|
||||
return localStorage.GetStreamFieldNames(ctx, tenantIDs, q)
|
||||
}
|
||||
return netstorageSelect.GetStreamFieldNames(ctx, tenantIDs, q)
|
||||
return strg.GetStreamFieldNames(ctx, tenantIDs, q)
|
||||
}
|
||||
|
||||
// GetStreamFieldValues executes q and returns stream field values for the given fieldName seen in results.
|
||||
//
|
||||
// If limit > 0, then up to limit unique stream field values are returned.
|
||||
func GetStreamFieldValues(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, fieldName string, limit uint64) ([]logstorage.ValueWithHits, error) {
|
||||
if localStorage != nil {
|
||||
return localStorage.GetStreamFieldValues(ctx, tenantIDs, q, fieldName, limit)
|
||||
}
|
||||
return netstorageSelect.GetStreamFieldValues(ctx, tenantIDs, q, fieldName, limit)
|
||||
return strg.GetStreamFieldValues(ctx, tenantIDs, q, fieldName, limit)
|
||||
}
|
||||
|
||||
// GetStreams executes q and returns streams seen in query results.
|
||||
//
|
||||
// If limit > 0, then up to limit unique streams are returned.
|
||||
func GetStreams(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, limit uint64) ([]logstorage.ValueWithHits, error) {
|
||||
if localStorage != nil {
|
||||
return localStorage.GetStreams(ctx, tenantIDs, q, limit)
|
||||
}
|
||||
return netstorageSelect.GetStreams(ctx, tenantIDs, q, limit)
|
||||
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) {
|
||||
if localStorage != nil {
|
||||
return localStorage.GetStreamIDs(ctx, tenantIDs, q, limit)
|
||||
}
|
||||
return netstorageSelect.GetStreamIDs(ctx, tenantIDs, q, limit)
|
||||
return strg.GetStreamIDs(ctx, tenantIDs, q, limit)
|
||||
}
|
||||
|
||||
func writeStorageMetrics(w io.Writer, strg *logstorage.Storage) {
|
||||
|
||||
@@ -1,369 +0,0 @@
|
||||
package netinsert
|
||||
|
||||
import (
|
||||
"errors"
|
||||
"fmt"
|
||||
"io"
|
||||
"net/http"
|
||||
"net/url"
|
||||
"sync"
|
||||
"sync/atomic"
|
||||
"time"
|
||||
|
||||
"github.com/valyala/fastrand"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/contextutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/encoding/zstd"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/fasttime"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/promauth"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/timerpool"
|
||||
)
|
||||
|
||||
// the maximum size of a single data block sent to storage node.
|
||||
const maxInsertBlockSize = 2 * 1024 * 1024
|
||||
|
||||
// ProtocolVersion is the version of the data ingestion protocol.
|
||||
//
|
||||
// It must be changed every time the data encoding at /internal/insert HTTP endpoint is changed.
|
||||
const ProtocolVersion = "v1"
|
||||
|
||||
// Storage is a network storage for sending data to remote storage nodes in the cluster.
|
||||
type Storage struct {
|
||||
sns []*storageNode
|
||||
|
||||
disableCompression bool
|
||||
|
||||
srt *streamRowsTracker
|
||||
|
||||
pendingDataBuffers chan *bytesutil.ByteBuffer
|
||||
|
||||
stopCh chan struct{}
|
||||
wg sync.WaitGroup
|
||||
}
|
||||
|
||||
type storageNode struct {
|
||||
// scheme is http or https scheme to communicate with addr
|
||||
scheme string
|
||||
|
||||
// addr is TCP address of storage node to send the ingested data to
|
||||
addr string
|
||||
|
||||
// s is a storage, which holds the given storageNode
|
||||
s *Storage
|
||||
|
||||
// c is an http client used for sending data blocks to addr.
|
||||
c *http.Client
|
||||
|
||||
// ac is auth config used for setting request headers such as Authorization and Host.
|
||||
ac *promauth.Config
|
||||
|
||||
// pendingData contains pending data, which must be sent to the storage node at the addr.
|
||||
pendingDataMu sync.Mutex
|
||||
pendingData *bytesutil.ByteBuffer
|
||||
pendingDataLastFlush time.Time
|
||||
|
||||
// the unix timestamp until the storageNode is disabled for data writing.
|
||||
disabledUntil atomic.Uint64
|
||||
}
|
||||
|
||||
func newStorageNode(s *Storage, addr string, ac *promauth.Config, isTLS bool) *storageNode {
|
||||
tr := httputil.NewTransport(false, "vlinsert_backend")
|
||||
tr.TLSHandshakeTimeout = 20 * time.Second
|
||||
tr.DisableCompression = true
|
||||
|
||||
scheme := "http"
|
||||
if isTLS {
|
||||
scheme = "https"
|
||||
}
|
||||
|
||||
sn := &storageNode{
|
||||
scheme: scheme,
|
||||
addr: addr,
|
||||
s: s,
|
||||
c: &http.Client{
|
||||
Transport: ac.NewRoundTripper(tr),
|
||||
},
|
||||
ac: ac,
|
||||
|
||||
pendingData: &bytesutil.ByteBuffer{},
|
||||
}
|
||||
|
||||
s.wg.Add(1)
|
||||
go func() {
|
||||
defer s.wg.Done()
|
||||
sn.backgroundFlusher()
|
||||
}()
|
||||
|
||||
return sn
|
||||
}
|
||||
|
||||
func (sn *storageNode) backgroundFlusher() {
|
||||
t := time.NewTicker(time.Second)
|
||||
defer t.Stop()
|
||||
|
||||
for {
|
||||
select {
|
||||
case <-sn.s.stopCh:
|
||||
return
|
||||
case <-t.C:
|
||||
sn.flushPendingData()
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func (sn *storageNode) flushPendingData() {
|
||||
sn.pendingDataMu.Lock()
|
||||
if time.Since(sn.pendingDataLastFlush) < time.Second {
|
||||
// nothing to flush
|
||||
sn.pendingDataMu.Unlock()
|
||||
return
|
||||
}
|
||||
|
||||
pendingData := sn.grabPendingDataForFlushLocked()
|
||||
sn.pendingDataMu.Unlock()
|
||||
|
||||
sn.mustSendInsertRequest(pendingData)
|
||||
}
|
||||
|
||||
func (sn *storageNode) addRow(r *logstorage.InsertRow) {
|
||||
bb := bbPool.Get()
|
||||
b := bb.B
|
||||
|
||||
b = r.Marshal(b)
|
||||
|
||||
if len(b) > maxInsertBlockSize {
|
||||
logger.Warnf("skipping too long log entry, since its length exceeds %d bytes; the actual log entry length is %d bytes; log entry contents: %s", maxInsertBlockSize, len(b), b)
|
||||
return
|
||||
}
|
||||
|
||||
var pendingData *bytesutil.ByteBuffer
|
||||
sn.pendingDataMu.Lock()
|
||||
if sn.pendingData.Len()+len(b) > maxInsertBlockSize {
|
||||
pendingData = sn.grabPendingDataForFlushLocked()
|
||||
}
|
||||
sn.pendingData.MustWrite(b)
|
||||
sn.pendingDataMu.Unlock()
|
||||
|
||||
bb.B = b
|
||||
bbPool.Put(bb)
|
||||
|
||||
if pendingData != nil {
|
||||
sn.mustSendInsertRequest(pendingData)
|
||||
}
|
||||
}
|
||||
|
||||
var bbPool bytesutil.ByteBufferPool
|
||||
|
||||
func (sn *storageNode) grabPendingDataForFlushLocked() *bytesutil.ByteBuffer {
|
||||
sn.pendingDataLastFlush = time.Now()
|
||||
pendingData := sn.pendingData
|
||||
sn.pendingData = <-sn.s.pendingDataBuffers
|
||||
|
||||
return pendingData
|
||||
}
|
||||
|
||||
func (sn *storageNode) mustSendInsertRequest(pendingData *bytesutil.ByteBuffer) {
|
||||
defer func() {
|
||||
pendingData.Reset()
|
||||
sn.s.pendingDataBuffers <- pendingData
|
||||
}()
|
||||
|
||||
err := sn.sendInsertRequest(pendingData)
|
||||
if err == nil {
|
||||
return
|
||||
}
|
||||
|
||||
if !errors.Is(err, errTemporarilyDisabled) {
|
||||
logger.Warnf("%s; re-routing the data block to the remaining nodes", err)
|
||||
}
|
||||
for !sn.s.sendInsertRequestToAnyNode(pendingData) {
|
||||
logger.Errorf("cannot send pending data to all storage nodes, since all of them are unavailable; re-trying to send the data in a second")
|
||||
|
||||
t := timerpool.Get(time.Second)
|
||||
select {
|
||||
case <-sn.s.stopCh:
|
||||
timerpool.Put(t)
|
||||
logger.Errorf("dropping %d bytes of data, since there are no available storage nodes", pendingData.Len())
|
||||
return
|
||||
case <-t.C:
|
||||
timerpool.Put(t)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func (sn *storageNode) sendInsertRequest(pendingData *bytesutil.ByteBuffer) error {
|
||||
dataLen := pendingData.Len()
|
||||
if dataLen == 0 {
|
||||
// Nothing to send.
|
||||
return nil
|
||||
}
|
||||
|
||||
if sn.disabledUntil.Load() > fasttime.UnixTimestamp() {
|
||||
return errTemporarilyDisabled
|
||||
}
|
||||
|
||||
ctx, cancel := contextutil.NewStopChanContext(sn.s.stopCh)
|
||||
defer cancel()
|
||||
|
||||
var body io.Reader
|
||||
if !sn.s.disableCompression {
|
||||
bb := zstdBufPool.Get()
|
||||
defer zstdBufPool.Put(bb)
|
||||
|
||||
bb.B = zstd.CompressLevel(bb.B[:0], pendingData.B, 1)
|
||||
body = bb.NewReader()
|
||||
} else {
|
||||
body = pendingData.NewReader()
|
||||
}
|
||||
|
||||
reqURL := sn.getRequestURL("/internal/insert")
|
||||
req, err := http.NewRequestWithContext(ctx, "POST", reqURL, body)
|
||||
if err != nil {
|
||||
logger.Panicf("BUG: unexpected error when creating an http request: %s", err)
|
||||
}
|
||||
req.Header.Set("Content-Type", "application/octet-stream")
|
||||
if !sn.s.disableCompression {
|
||||
req.Header.Set("Content-Encoding", "zstd")
|
||||
}
|
||||
if err := sn.ac.SetHeaders(req, true); err != nil {
|
||||
return fmt.Errorf("cannot set auth headers for %q: %w", reqURL, err)
|
||||
}
|
||||
|
||||
resp, err := sn.c.Do(req)
|
||||
if err != nil {
|
||||
// Disable sn for data writing for 10 seconds.
|
||||
sn.disabledUntil.Store(fasttime.UnixTimestamp() + 10)
|
||||
|
||||
return fmt.Errorf("cannot send data block with the length %d to %q: %s", pendingData.Len(), reqURL, err)
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
|
||||
if resp.StatusCode/100 == 2 {
|
||||
return nil
|
||||
}
|
||||
|
||||
respBody, err := io.ReadAll(resp.Body)
|
||||
if err != nil {
|
||||
respBody = []byte(fmt.Sprintf("%s", err))
|
||||
}
|
||||
|
||||
// Disable sn for data writing for 10 seconds.
|
||||
sn.disabledUntil.Store(fasttime.UnixTimestamp() + 10)
|
||||
|
||||
return fmt.Errorf("unexpected status code returned when sending data block to %q: %d; want 2xx; response body: %q", reqURL, resp.StatusCode, respBody)
|
||||
}
|
||||
|
||||
func (sn *storageNode) getRequestURL(path string) string {
|
||||
return fmt.Sprintf("%s://%s%s?version=%s", sn.scheme, sn.addr, path, url.QueryEscape(ProtocolVersion))
|
||||
}
|
||||
|
||||
var zstdBufPool bytesutil.ByteBufferPool
|
||||
|
||||
// NewStorage returns new Storage for the given addrs with the given authCfgs.
|
||||
//
|
||||
// The concurrency is the average number of concurrent connections per every addr.
|
||||
//
|
||||
// If disableCompression is set, then the data is sent uncompressed to the remote storage.
|
||||
//
|
||||
// Call MustStop on the returned storage when it is no longer needed.
|
||||
func NewStorage(addrs []string, authCfgs []*promauth.Config, isTLSs []bool, concurrency int, disableCompression bool) *Storage {
|
||||
pendingDataBuffers := make(chan *bytesutil.ByteBuffer, concurrency*len(addrs))
|
||||
for i := 0; i < cap(pendingDataBuffers); i++ {
|
||||
pendingDataBuffers <- &bytesutil.ByteBuffer{}
|
||||
}
|
||||
|
||||
s := &Storage{
|
||||
disableCompression: disableCompression,
|
||||
pendingDataBuffers: pendingDataBuffers,
|
||||
stopCh: make(chan struct{}),
|
||||
}
|
||||
|
||||
sns := make([]*storageNode, len(addrs))
|
||||
for i, addr := range addrs {
|
||||
sns[i] = newStorageNode(s, addr, authCfgs[i], isTLSs[i])
|
||||
}
|
||||
s.sns = sns
|
||||
|
||||
s.srt = newStreamRowsTracker(len(sns))
|
||||
|
||||
return s
|
||||
}
|
||||
|
||||
// MustStop stops the s.
|
||||
func (s *Storage) MustStop() {
|
||||
close(s.stopCh)
|
||||
s.wg.Wait()
|
||||
s.sns = nil
|
||||
}
|
||||
|
||||
// AddRow adds the given log row into s.
|
||||
func (s *Storage) AddRow(streamHash uint64, r *logstorage.InsertRow) {
|
||||
idx := s.srt.getNodeIdx(streamHash)
|
||||
sn := s.sns[idx]
|
||||
sn.addRow(r)
|
||||
}
|
||||
|
||||
func (s *Storage) sendInsertRequestToAnyNode(pendingData *bytesutil.ByteBuffer) bool {
|
||||
startIdx := int(fastrand.Uint32n(uint32(len(s.sns))))
|
||||
for i := range s.sns {
|
||||
idx := (startIdx + i) % len(s.sns)
|
||||
sn := s.sns[idx]
|
||||
err := sn.sendInsertRequest(pendingData)
|
||||
if err == nil {
|
||||
return true
|
||||
}
|
||||
if !errors.Is(err, errTemporarilyDisabled) {
|
||||
logger.Warnf("cannot send pending data to the storage node %q: %s; trying to send it to another storage node", sn.addr, err)
|
||||
}
|
||||
}
|
||||
return false
|
||||
}
|
||||
|
||||
var errTemporarilyDisabled = fmt.Errorf("writing to the node is temporarily disabled")
|
||||
|
||||
type streamRowsTracker struct {
|
||||
mu sync.Mutex
|
||||
|
||||
nodesCount int64
|
||||
rowsPerStream map[uint64]uint64
|
||||
}
|
||||
|
||||
func newStreamRowsTracker(nodesCount int) *streamRowsTracker {
|
||||
return &streamRowsTracker{
|
||||
nodesCount: int64(nodesCount),
|
||||
rowsPerStream: make(map[uint64]uint64),
|
||||
}
|
||||
}
|
||||
|
||||
func (srt *streamRowsTracker) getNodeIdx(streamHash uint64) uint64 {
|
||||
if srt.nodesCount == 1 {
|
||||
// Fast path for a single node.
|
||||
return 0
|
||||
}
|
||||
|
||||
srt.mu.Lock()
|
||||
defer srt.mu.Unlock()
|
||||
|
||||
streamRows := srt.rowsPerStream[streamHash] + 1
|
||||
srt.rowsPerStream[streamHash] = streamRows
|
||||
|
||||
if streamRows <= 1000 {
|
||||
// Write the initial rows for the stream to a single storage node for better locality.
|
||||
// This should work great for log streams containing small number of logs, since will be distributed
|
||||
// evenly among available storage nodes because they have different streamHash.
|
||||
return streamHash % uint64(srt.nodesCount)
|
||||
}
|
||||
|
||||
// The log stream contains more than 1000 rows. Distribute them among storage nodes at random
|
||||
// in order to improve query performance over this stream (the data for the log stream
|
||||
// can be processed in parallel on all the storage nodes).
|
||||
//
|
||||
// The random distribution is preferred over round-robin distribution in order to avoid possible
|
||||
// dependency between the order of the ingested logs and the number of storage nodes,
|
||||
// which may lead to non-uniform distribution of logs among storage nodes.
|
||||
return uint64(fastrand.Uint32n(uint32(srt.nodesCount)))
|
||||
}
|
||||
@@ -1,57 +0,0 @@
|
||||
package netinsert
|
||||
|
||||
import (
|
||||
"fmt"
|
||||
"math"
|
||||
"math/rand"
|
||||
"testing"
|
||||
|
||||
"github.com/cespare/xxhash/v2"
|
||||
)
|
||||
|
||||
func TestStreamRowsTracker(t *testing.T) {
|
||||
f := func(rowsCount, streamsCount, nodesCount int) {
|
||||
t.Helper()
|
||||
|
||||
// generate stream hashes
|
||||
streamHashes := make([]uint64, streamsCount)
|
||||
for i := range streamHashes {
|
||||
streamHashes[i] = xxhash.Sum64([]byte(fmt.Sprintf("stream %d.", i)))
|
||||
}
|
||||
|
||||
srt := newStreamRowsTracker(nodesCount)
|
||||
|
||||
rng := rand.New(rand.NewSource(0))
|
||||
rowsPerNode := make([]uint64, nodesCount)
|
||||
for i := 0; i < rowsCount; i++ {
|
||||
streamIdx := rng.Intn(streamsCount)
|
||||
h := streamHashes[streamIdx]
|
||||
nodeIdx := srt.getNodeIdx(h)
|
||||
rowsPerNode[nodeIdx]++
|
||||
}
|
||||
|
||||
// Verify that rows are uniformly distributed among nodes.
|
||||
expectedRowsPerNode := float64(rowsCount) / float64(nodesCount)
|
||||
for nodeIdx, nodeRows := range rowsPerNode {
|
||||
if math.Abs(float64(nodeRows)-expectedRowsPerNode)/expectedRowsPerNode > 0.15 {
|
||||
t.Fatalf("non-uniform distribution of rows among nodes; node %d has %d rows, while it must have %v rows; rowsPerNode=%d",
|
||||
nodeIdx, nodeRows, expectedRowsPerNode, rowsPerNode)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
rowsCount := 10000
|
||||
streamsCount := 9
|
||||
nodesCount := 2
|
||||
f(rowsCount, streamsCount, nodesCount)
|
||||
|
||||
rowsCount = 10000
|
||||
streamsCount = 100
|
||||
nodesCount = 2
|
||||
f(rowsCount, streamsCount, nodesCount)
|
||||
|
||||
rowsCount = 100000
|
||||
streamsCount = 1000
|
||||
nodesCount = 9
|
||||
f(rowsCount, streamsCount, nodesCount)
|
||||
}
|
||||
@@ -1,472 +0,0 @@
|
||||
package netselect
|
||||
|
||||
import (
|
||||
"context"
|
||||
"errors"
|
||||
"fmt"
|
||||
"io"
|
||||
"math"
|
||||
"net/http"
|
||||
"net/url"
|
||||
"strings"
|
||||
"sync"
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/contextutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/encoding"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/encoding/zstd"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/promauth"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/slicesutil"
|
||||
)
|
||||
|
||||
const (
|
||||
// FieldNamesProtocolVersion is the version of the protocol used for /internal/select/field_names HTTP endpoint.
|
||||
//
|
||||
// It must be updated every time the protocol changes.
|
||||
FieldNamesProtocolVersion = "v1"
|
||||
|
||||
// FieldValuesProtocolVersion is the version of the protocol used for /internal/select/field_values HTTP endpoint.
|
||||
//
|
||||
// It must be updated every time the protocol changes.
|
||||
FieldValuesProtocolVersion = "v1"
|
||||
|
||||
// StreamFieldNamesProtocolVersion is the version of the protocol used for /internal/select/stream_field_names HTTP endpoint.
|
||||
//
|
||||
// It must be updated every time the protocol changes.
|
||||
StreamFieldNamesProtocolVersion = "v1"
|
||||
|
||||
// StreamFieldValuesProtocolVersion is the version of the protocol used for /internal/select/stream_field_values HTTP endpoint.
|
||||
//
|
||||
// It must be updated every time the protocol changes.
|
||||
StreamFieldValuesProtocolVersion = "v1"
|
||||
|
||||
// StreamsProtocolVersion is the version of the protocol used for /internal/select/streams HTTP endpoint.
|
||||
//
|
||||
// It must be updated every time the protocol changes.
|
||||
StreamsProtocolVersion = "v1"
|
||||
|
||||
// StreamIDsProtocolVersion is the version of the protocol used for /internal/select/stream_ids HTTP endpoint.
|
||||
//
|
||||
// It must be updated every time the protocol changes.
|
||||
StreamIDsProtocolVersion = "v1"
|
||||
|
||||
// QueryProtocolVersion is the version of the protocol used for /internal/select/query HTTP endpoint.
|
||||
//
|
||||
// It must be updated every time the protocol changes.
|
||||
QueryProtocolVersion = "v1"
|
||||
)
|
||||
|
||||
// Storage is a network storage for querying remote storage nodes in the cluster.
|
||||
type Storage struct {
|
||||
sns []*storageNode
|
||||
|
||||
disableCompression bool
|
||||
}
|
||||
|
||||
type storageNode struct {
|
||||
// scheme is http or https scheme to communicate with addr
|
||||
scheme string
|
||||
|
||||
// addr is TCP address of the storage node to query
|
||||
addr string
|
||||
|
||||
// s is a storage, which holds the given storageNode
|
||||
s *Storage
|
||||
|
||||
// c is an http client used for querying storage node at addr.
|
||||
c *http.Client
|
||||
|
||||
// ac is auth config used for setting request headers such as Authorization and Host.
|
||||
ac *promauth.Config
|
||||
}
|
||||
|
||||
func newStorageNode(s *Storage, addr string, ac *promauth.Config, isTLS bool) *storageNode {
|
||||
tr := httputil.NewTransport(false, "vlselect_backend")
|
||||
tr.TLSHandshakeTimeout = 20 * time.Second
|
||||
tr.DisableCompression = true
|
||||
|
||||
scheme := "http"
|
||||
if isTLS {
|
||||
scheme = "https"
|
||||
}
|
||||
|
||||
sn := &storageNode{
|
||||
scheme: scheme,
|
||||
addr: addr,
|
||||
s: s,
|
||||
c: &http.Client{
|
||||
Transport: ac.NewRoundTripper(tr),
|
||||
},
|
||||
ac: ac,
|
||||
}
|
||||
return sn
|
||||
}
|
||||
|
||||
func (sn *storageNode) runQuery(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, processBlock func(db *logstorage.DataBlock)) error {
|
||||
args := sn.getCommonArgs(QueryProtocolVersion, tenantIDs, q)
|
||||
|
||||
reqURL := sn.getRequestURL("/internal/select/query", args)
|
||||
req, err := http.NewRequestWithContext(ctx, "GET", reqURL, nil)
|
||||
if err != nil {
|
||||
logger.Panicf("BUG: unexpected error when creating a request: %s", err)
|
||||
}
|
||||
if err := sn.ac.SetHeaders(req, true); err != nil {
|
||||
return fmt.Errorf("cannot set auth headers for %q: %w", reqURL, err)
|
||||
}
|
||||
|
||||
// send the request to the storage node
|
||||
resp, err := sn.c.Do(req)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
|
||||
if resp.StatusCode != http.StatusOK {
|
||||
responseBody, err := io.ReadAll(resp.Body)
|
||||
if err != nil {
|
||||
responseBody = []byte(err.Error())
|
||||
}
|
||||
return fmt.Errorf("unexpected status code for the request to %q: %d; want %d; response: %q", reqURL, resp.StatusCode, http.StatusOK, responseBody)
|
||||
}
|
||||
|
||||
// read the response
|
||||
var dataLenBuf [8]byte
|
||||
var buf []byte
|
||||
var db logstorage.DataBlock
|
||||
var valuesBuf []string
|
||||
for {
|
||||
if _, err := io.ReadFull(resp.Body, dataLenBuf[:]); err != nil {
|
||||
if errors.Is(err, io.EOF) {
|
||||
// The end of response stream
|
||||
return nil
|
||||
}
|
||||
return fmt.Errorf("cannot read block size from %q: %w", reqURL, err)
|
||||
}
|
||||
blockLen := encoding.UnmarshalUint64(dataLenBuf[:])
|
||||
if blockLen > math.MaxInt {
|
||||
return fmt.Errorf("too big data block: %d bytes; mustn't exceed %v bytes", blockLen, math.MaxInt)
|
||||
}
|
||||
|
||||
buf = slicesutil.SetLength(buf, int(blockLen))
|
||||
if _, err := io.ReadFull(resp.Body, buf); err != nil {
|
||||
return fmt.Errorf("cannot read block with size of %d bytes from %q: %w", blockLen, reqURL, err)
|
||||
}
|
||||
|
||||
src := buf
|
||||
if !sn.s.disableCompression {
|
||||
bufLen := len(buf)
|
||||
var err error
|
||||
buf, err = zstd.Decompress(buf, buf)
|
||||
if err != nil {
|
||||
return fmt.Errorf("cannot decompress data block: %w", err)
|
||||
}
|
||||
src = buf[bufLen:]
|
||||
}
|
||||
|
||||
for len(src) > 0 {
|
||||
tail, vb, err := db.UnmarshalInplace(src, valuesBuf[:0])
|
||||
if err != nil {
|
||||
return fmt.Errorf("cannot unmarshal data block received from %q: %w", reqURL, err)
|
||||
}
|
||||
valuesBuf = vb
|
||||
src = tail
|
||||
|
||||
processBlock(&db)
|
||||
|
||||
clear(valuesBuf)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func (sn *storageNode) getFieldNames(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query) ([]logstorage.ValueWithHits, error) {
|
||||
args := sn.getCommonArgs(FieldNamesProtocolVersion, tenantIDs, q)
|
||||
|
||||
return sn.getValuesWithHits(ctx, "/internal/select/field_names", args)
|
||||
}
|
||||
|
||||
func (sn *storageNode) getFieldValues(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, fieldName string, limit uint64) ([]logstorage.ValueWithHits, error) {
|
||||
args := sn.getCommonArgs(FieldValuesProtocolVersion, tenantIDs, q)
|
||||
args.Set("field", fieldName)
|
||||
args.Set("limit", fmt.Sprintf("%d", limit))
|
||||
|
||||
return sn.getValuesWithHits(ctx, "/internal/select/field_values", args)
|
||||
}
|
||||
|
||||
func (sn *storageNode) getStreamFieldNames(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query) ([]logstorage.ValueWithHits, error) {
|
||||
args := sn.getCommonArgs(StreamFieldNamesProtocolVersion, tenantIDs, q)
|
||||
|
||||
return sn.getValuesWithHits(ctx, "/internal/select/stream_field_names", args)
|
||||
}
|
||||
|
||||
func (sn *storageNode) getStreamFieldValues(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, fieldName string, limit uint64) ([]logstorage.ValueWithHits, error) {
|
||||
args := sn.getCommonArgs(StreamFieldValuesProtocolVersion, tenantIDs, q)
|
||||
args.Set("field", fieldName)
|
||||
args.Set("limit", fmt.Sprintf("%d", limit))
|
||||
|
||||
return sn.getValuesWithHits(ctx, "/internal/select/stream_field_values", args)
|
||||
}
|
||||
|
||||
func (sn *storageNode) getStreams(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, limit uint64) ([]logstorage.ValueWithHits, error) {
|
||||
args := sn.getCommonArgs(StreamsProtocolVersion, tenantIDs, q)
|
||||
args.Set("limit", fmt.Sprintf("%d", limit))
|
||||
|
||||
return sn.getValuesWithHits(ctx, "/internal/select/streams", args)
|
||||
}
|
||||
|
||||
func (sn *storageNode) getStreamIDs(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, limit uint64) ([]logstorage.ValueWithHits, error) {
|
||||
args := sn.getCommonArgs(StreamIDsProtocolVersion, tenantIDs, q)
|
||||
args.Set("limit", fmt.Sprintf("%d", limit))
|
||||
|
||||
return sn.getValuesWithHits(ctx, "/internal/select/stream_ids", args)
|
||||
}
|
||||
|
||||
func (sn *storageNode) getCommonArgs(version string, tenantIDs []logstorage.TenantID, q *logstorage.Query) url.Values {
|
||||
args := url.Values{}
|
||||
args.Set("version", version)
|
||||
args.Set("tenant_ids", string(logstorage.MarshalTenantIDs(nil, tenantIDs)))
|
||||
args.Set("query", q.String())
|
||||
args.Set("timestamp", fmt.Sprintf("%d", q.GetTimestamp()))
|
||||
args.Set("disable_compression", fmt.Sprintf("%v", sn.s.disableCompression))
|
||||
return args
|
||||
}
|
||||
|
||||
func (sn *storageNode) getValuesWithHits(ctx context.Context, path string, args url.Values) ([]logstorage.ValueWithHits, error) {
|
||||
data, err := sn.executeRequestAt(ctx, path, args)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
return unmarshalValuesWithHits(data)
|
||||
}
|
||||
|
||||
func (sn *storageNode) executeRequestAt(ctx context.Context, path string, args url.Values) ([]byte, error) {
|
||||
reqURL := sn.getRequestURL(path, args)
|
||||
req, err := http.NewRequestWithContext(ctx, "GET", reqURL, nil)
|
||||
if err != nil {
|
||||
logger.Panicf("BUG: unexpected error when creating a request: %s", err)
|
||||
}
|
||||
if err := sn.ac.SetHeaders(req, true); err != nil {
|
||||
return nil, fmt.Errorf("cannot set auth headers for %q: %w", reqURL, err)
|
||||
}
|
||||
|
||||
// send the request to the storage node
|
||||
resp, err := sn.c.Do(req)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
defer resp.Body.Close()
|
||||
|
||||
if resp.StatusCode != http.StatusOK {
|
||||
responseBody, err := io.ReadAll(resp.Body)
|
||||
if err != nil {
|
||||
responseBody = []byte(err.Error())
|
||||
}
|
||||
return nil, fmt.Errorf("unexpected status code for the request to %q: %d; want %d; response: %q", reqURL, resp.StatusCode, http.StatusOK, responseBody)
|
||||
}
|
||||
|
||||
// read the response
|
||||
var bb bytesutil.ByteBuffer
|
||||
if _, err := bb.ReadFrom(resp.Body); err != nil {
|
||||
return nil, fmt.Errorf("cannot read response from %q: %w", reqURL, err)
|
||||
}
|
||||
|
||||
if sn.s.disableCompression {
|
||||
return bb.B, nil
|
||||
}
|
||||
|
||||
bbLen := len(bb.B)
|
||||
bb.B, err = zstd.Decompress(bb.B, bb.B)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
return bb.B[bbLen:], nil
|
||||
}
|
||||
|
||||
func (sn *storageNode) getRequestURL(path string, args url.Values) string {
|
||||
return fmt.Sprintf("%s://%s%s?%s", sn.scheme, sn.addr, path, args.Encode())
|
||||
}
|
||||
|
||||
// NewStorage returns new Storage for the given addrs and the given authCfgs.
|
||||
//
|
||||
// If disableCompression is set, then uncompressed responses are received from storage nodes.
|
||||
//
|
||||
// Call MustStop on the returned storage when it is no longer needed.
|
||||
func NewStorage(addrs []string, authCfgs []*promauth.Config, isTLSs []bool, disableCompression bool) *Storage {
|
||||
s := &Storage{
|
||||
disableCompression: disableCompression,
|
||||
}
|
||||
|
||||
sns := make([]*storageNode, len(addrs))
|
||||
for i, addr := range addrs {
|
||||
sns[i] = newStorageNode(s, addr, authCfgs[i], isTLSs[i])
|
||||
}
|
||||
s.sns = sns
|
||||
|
||||
return s
|
||||
}
|
||||
|
||||
// MustStop stops the s.
|
||||
func (s *Storage) MustStop() {
|
||||
s.sns = nil
|
||||
}
|
||||
|
||||
// RunQuery runs the given q and calls writeBlock for the returned data blocks
|
||||
func (s *Storage) RunQuery(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, writeBlock logstorage.WriteDataBlockFunc) error {
|
||||
nqr, err := logstorage.NewNetQueryRunner(ctx, tenantIDs, q, s.RunQuery, writeBlock)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
search := func(stopCh <-chan struct{}, q *logstorage.Query, writeBlock logstorage.WriteDataBlockFunc) error {
|
||||
return s.runQuery(stopCh, tenantIDs, q, writeBlock)
|
||||
}
|
||||
|
||||
concurrency := q.GetConcurrency()
|
||||
return nqr.Run(ctx, concurrency, search)
|
||||
}
|
||||
|
||||
func (s *Storage) runQuery(stopCh <-chan struct{}, tenantIDs []logstorage.TenantID, q *logstorage.Query, writeBlock logstorage.WriteDataBlockFunc) error {
|
||||
ctxWithCancel, cancel := contextutil.NewStopChanContext(stopCh)
|
||||
defer cancel()
|
||||
|
||||
errs := make([]error, len(s.sns))
|
||||
|
||||
var wg sync.WaitGroup
|
||||
for i := range s.sns {
|
||||
wg.Add(1)
|
||||
go func(nodeIdx int) {
|
||||
defer wg.Done()
|
||||
sn := s.sns[nodeIdx]
|
||||
err := sn.runQuery(ctxWithCancel, tenantIDs, q, func(db *logstorage.DataBlock) {
|
||||
writeBlock(uint(nodeIdx), db)
|
||||
})
|
||||
if err != nil {
|
||||
// Cancel the remaining parallel queries
|
||||
cancel()
|
||||
}
|
||||
|
||||
errs[nodeIdx] = err
|
||||
}(i)
|
||||
}
|
||||
wg.Wait()
|
||||
|
||||
return getFirstNonCancelError(errs)
|
||||
}
|
||||
|
||||
// GetFieldNames executes q and returns field names seen in results.
|
||||
func (s *Storage) GetFieldNames(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query) ([]logstorage.ValueWithHits, error) {
|
||||
return s.getValuesWithHits(ctx, 0, false, func(ctx context.Context, sn *storageNode) ([]logstorage.ValueWithHits, error) {
|
||||
return sn.getFieldNames(ctx, tenantIDs, q)
|
||||
})
|
||||
}
|
||||
|
||||
// GetFieldValues executes q and returns unique values for the fieldName seen in results.
|
||||
//
|
||||
// If limit > 0, then up to limit unique values are returned.
|
||||
func (s *Storage) GetFieldValues(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, fieldName string, limit uint64) ([]logstorage.ValueWithHits, error) {
|
||||
return s.getValuesWithHits(ctx, limit, true, func(ctx context.Context, sn *storageNode) ([]logstorage.ValueWithHits, error) {
|
||||
return sn.getFieldValues(ctx, tenantIDs, q, fieldName, limit)
|
||||
})
|
||||
}
|
||||
|
||||
// GetStreamFieldNames executes q and returns stream field names seen in results.
|
||||
func (s *Storage) GetStreamFieldNames(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query) ([]logstorage.ValueWithHits, error) {
|
||||
return s.getValuesWithHits(ctx, 0, false, func(ctx context.Context, sn *storageNode) ([]logstorage.ValueWithHits, error) {
|
||||
return sn.getStreamFieldNames(ctx, tenantIDs, q)
|
||||
})
|
||||
}
|
||||
|
||||
// GetStreamFieldValues executes q and returns stream field values for the given fieldName seen in results.
|
||||
//
|
||||
// If limit > 0, then up to limit unique stream field values are returned.
|
||||
func (s *Storage) GetStreamFieldValues(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, fieldName string, limit uint64) ([]logstorage.ValueWithHits, error) {
|
||||
return s.getValuesWithHits(ctx, limit, true, func(ctx context.Context, sn *storageNode) ([]logstorage.ValueWithHits, error) {
|
||||
return sn.getStreamFieldValues(ctx, tenantIDs, q, fieldName, limit)
|
||||
})
|
||||
}
|
||||
|
||||
// GetStreams executes q and returns streams seen in query results.
|
||||
//
|
||||
// If limit > 0, then up to limit unique streams are returned.
|
||||
func (s *Storage) GetStreams(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, limit uint64) ([]logstorage.ValueWithHits, error) {
|
||||
return s.getValuesWithHits(ctx, limit, true, func(ctx context.Context, sn *storageNode) ([]logstorage.ValueWithHits, error) {
|
||||
return sn.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 (s *Storage) GetStreamIDs(ctx context.Context, tenantIDs []logstorage.TenantID, q *logstorage.Query, limit uint64) ([]logstorage.ValueWithHits, error) {
|
||||
return s.getValuesWithHits(ctx, limit, true, func(ctx context.Context, sn *storageNode) ([]logstorage.ValueWithHits, error) {
|
||||
return sn.getStreamIDs(ctx, tenantIDs, q, limit)
|
||||
})
|
||||
}
|
||||
|
||||
func (s *Storage) getValuesWithHits(ctx context.Context, limit uint64, resetHitsOnLimitExceeded bool,
|
||||
callback func(ctx context.Context, sn *storageNode) ([]logstorage.ValueWithHits, error)) ([]logstorage.ValueWithHits, error) {
|
||||
|
||||
ctxWithCancel, cancel := context.WithCancel(ctx)
|
||||
defer cancel()
|
||||
|
||||
results := make([][]logstorage.ValueWithHits, len(s.sns))
|
||||
errs := make([]error, len(s.sns))
|
||||
|
||||
var wg sync.WaitGroup
|
||||
for i := range s.sns {
|
||||
wg.Add(1)
|
||||
go func(nodeIdx int) {
|
||||
defer wg.Done()
|
||||
|
||||
sn := s.sns[nodeIdx]
|
||||
vhs, err := callback(ctxWithCancel, sn)
|
||||
results[nodeIdx] = vhs
|
||||
errs[nodeIdx] = err
|
||||
|
||||
if err != nil {
|
||||
// Cancel the remaining parallel requests
|
||||
cancel()
|
||||
}
|
||||
}(i)
|
||||
}
|
||||
wg.Wait()
|
||||
|
||||
if err := getFirstNonCancelError(errs); err != nil {
|
||||
return nil, err
|
||||
}
|
||||
|
||||
vhs := logstorage.MergeValuesWithHits(results, limit, resetHitsOnLimitExceeded)
|
||||
|
||||
return vhs, nil
|
||||
}
|
||||
|
||||
func getFirstNonCancelError(errs []error) error {
|
||||
for _, err := range errs {
|
||||
if err != nil && !errors.Is(err, context.Canceled) {
|
||||
return err
|
||||
}
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
func unmarshalValuesWithHits(src []byte) ([]logstorage.ValueWithHits, error) {
|
||||
var vhs []logstorage.ValueWithHits
|
||||
for len(src) > 0 {
|
||||
var vh logstorage.ValueWithHits
|
||||
tail, err := vh.UnmarshalInplace(src)
|
||||
if err != nil {
|
||||
return nil, fmt.Errorf("cannot unmarshal ValueWithHits #%d: %w", len(vhs), err)
|
||||
}
|
||||
src = tail
|
||||
|
||||
// Clone vh.Value, since it points to src.
|
||||
vh.Value = strings.Clone(vh.Value)
|
||||
|
||||
vhs = append(vhs, vh)
|
||||
}
|
||||
|
||||
return vhs, nil
|
||||
}
|
||||
@@ -1,3 +1,3 @@
|
||||
See vmagent docs [here](https://docs.victoriametrics.com/victoriametrics/vmagent/).
|
||||
See vmagent docs [here](https://docs.victoriametrics.com/vmagent/).
|
||||
|
||||
vmagent docs can be edited at [docs/vmagent.md](https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/docs/victoriametrics/vmagent.md).
|
||||
vmagent docs can be edited at [docs/vmagent.md](https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/docs/vmagent.md).
|
||||
|
||||
@@ -105,7 +105,7 @@ func main() {
|
||||
// Some workloads may need increased GOGC values. Then such values can be set via GOGC environment variable.
|
||||
// It is recommended increasing GOGC if go_memstats_gc_cpu_fraction metric exposed at /metrics page
|
||||
// exceeds 0.05 for extended periods of time.
|
||||
cgroup.SetGOGC(50)
|
||||
cgroup.SetGOGC(30)
|
||||
|
||||
// Write flags and help message to stdout, since it is easier to grep or pipe.
|
||||
flag.CommandLine.SetOutput(os.Stdout)
|
||||
@@ -165,9 +165,7 @@ func main() {
|
||||
|
||||
promscrape.Init(remotewrite.PushDropSamplesOnFailure)
|
||||
|
||||
go httpserver.Serve(listenAddrs, requestHandler, httpserver.ServeOptions{
|
||||
UseProxyProtocol: useProxyProtocol,
|
||||
})
|
||||
go httpserver.Serve(listenAddrs, useProxyProtocol, requestHandler)
|
||||
logger.Infof("started vmagent in %.3f seconds", time.Since(startTime).Seconds())
|
||||
|
||||
pushmetrics.Init()
|
||||
@@ -244,7 +242,7 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
|
||||
}
|
||||
w.Header().Add("Content-Type", "text/html; charset=utf-8")
|
||||
fmt.Fprintf(w, "<h2>vmagent</h2>")
|
||||
fmt.Fprintf(w, "See docs at <a href='https://docs.victoriametrics.com/victoriametrics/vmagent/'>https://docs.victoriametrics.com/victoriametrics/vmagent/</a></br>")
|
||||
fmt.Fprintf(w, "See docs at <a href='https://docs.victoriametrics.com/vmagent/'>https://docs.victoriametrics.com/vmagent/</a></br>")
|
||||
fmt.Fprintf(w, "Useful endpoints:</br>")
|
||||
httpserver.WriteAPIHelp(w, [][2]string{
|
||||
{"targets", "status for discovered active targets"},
|
||||
@@ -445,10 +443,8 @@ func requestHandler(w http.ResponseWriter, r *http.Request) bool {
|
||||
case "/prometheus/api/v1/targets", "/api/v1/targets":
|
||||
promscrapeAPIV1TargetsRequests.Inc()
|
||||
w.Header().Set("Content-Type", "application/json")
|
||||
// https://prometheus.io/docs/prometheus/latest/querying/api/#targets
|
||||
state := r.FormValue("state")
|
||||
scrapePool := r.FormValue("scrapePool")
|
||||
promscrape.WriteAPIV1Targets(w, state, scrapePool)
|
||||
promscrape.WriteAPIV1Targets(w, state)
|
||||
return true
|
||||
case "/prometheus/target_response", "/target_response":
|
||||
promscrapeTargetResponseRequests.Inc()
|
||||
@@ -754,7 +750,7 @@ func usage() {
|
||||
const s = `
|
||||
vmagent collects metrics data via popular data ingestion protocols and routes it to VictoriaMetrics.
|
||||
|
||||
See the docs at https://docs.victoriametrics.com/victoriametrics/vmagent/ .
|
||||
See the docs at https://docs.victoriametrics.com/vmagent/ .
|
||||
`
|
||||
flagutil.Usage(s)
|
||||
}
|
||||
|
||||
@@ -9,5 +9,4 @@ COPY --from=certs /etc/ssl/certs/ca-certificates.crt /etc/ssl/certs/ca-certifica
|
||||
EXPOSE 8429
|
||||
ENTRYPOINT ["/vmagent-prod"]
|
||||
ARG TARGETARCH
|
||||
ARG BINARY_SUFFIX=non-existing
|
||||
COPY vmagent-linux-${TARGETARCH}-prod${BINARY_SUFFIX} ./vmagent-prod
|
||||
COPY vmagent-linux-${TARGETARCH}-prod ./vmagent-prod
|
||||
|
||||
@@ -10,29 +10,27 @@ import (
|
||||
"strconv"
|
||||
"strings"
|
||||
"sync"
|
||||
"sync/atomic"
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/awsapi"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/encoding"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/encoding/zstd"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httputil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/netutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/persistentqueue"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/promauth"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/protoparser/protoparserutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/ratelimiter"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/timerpool"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/timeutil"
|
||||
"github.com/VictoriaMetrics/metrics"
|
||||
"github.com/golang/snappy"
|
||||
)
|
||||
|
||||
var (
|
||||
forcePromProto = flagutil.NewArrayBool("remoteWrite.forcePromProto", "Whether to force Prometheus remote write protocol for sending data "+
|
||||
"to the corresponding -remoteWrite.url . See https://docs.victoriametrics.com/victoriametrics/vmagent/#victoriametrics-remote-write-protocol")
|
||||
"to the corresponding -remoteWrite.url . See https://docs.victoriametrics.com/vmagent/#victoriametrics-remote-write-protocol")
|
||||
forceVMProto = flagutil.NewArrayBool("remoteWrite.forceVMProto", "Whether to force VictoriaMetrics remote write protocol for sending data "+
|
||||
"to the corresponding -remoteWrite.url . See https://docs.victoriametrics.com/victoriametrics/vmagent/#victoriametrics-remote-write-protocol")
|
||||
"to the corresponding -remoteWrite.url . See https://docs.victoriametrics.com/vmagent/#victoriametrics-remote-write-protocol")
|
||||
|
||||
rateLimit = flagutil.NewArrayInt("remoteWrite.rateLimit", 0, "Optional rate limit in bytes per second for data sent to the corresponding -remoteWrite.url. "+
|
||||
"By default, the rate limit is disabled. It can be useful for limiting load on remote storage when big amounts of buffered data "+
|
||||
@@ -90,8 +88,7 @@ type client struct {
|
||||
remoteWriteURL string
|
||||
|
||||
// Whether to use VictoriaMetrics remote write protocol for sending the data to remoteWriteURL
|
||||
useVMProto atomic.Bool
|
||||
canDowngradeVMProto atomic.Bool
|
||||
useVMProto bool
|
||||
|
||||
fq *persistentqueue.FastQueue
|
||||
hc *http.Client
|
||||
@@ -129,7 +126,8 @@ func newHTTPClient(argIdx int, remoteWriteURL, sanitizedURL string, fq *persiste
|
||||
logger.Fatalf("cannot initialize AWS Config for -remoteWrite.url=%q: %s", remoteWriteURL, err)
|
||||
}
|
||||
|
||||
tr := httputil.NewTransport(false, "vmagent_remotewrite")
|
||||
tr := httputil.NewTransport(false)
|
||||
tr.DialContext = netutil.NewStatDialFunc("vmagent_remotewrite")
|
||||
tr.TLSHandshakeTimeout = tlsHandshakeTimeout.GetOptionalArg(argIdx)
|
||||
tr.MaxConnsPerHost = 2 * concurrency
|
||||
tr.MaxIdleConnsPerHost = 2 * concurrency
|
||||
@@ -170,11 +168,17 @@ func newHTTPClient(argIdx int, remoteWriteURL, sanitizedURL string, fq *persiste
|
||||
logger.Fatalf("-remoteWrite.useVMProto and -remoteWrite.usePromProto cannot be set simultaneously for -remoteWrite.url=%s", sanitizedURL)
|
||||
}
|
||||
if !useVMProto && !usePromProto {
|
||||
// The VM protocol could be downgraded later at runtime if unsupported media type response status is received.
|
||||
useVMProto = true
|
||||
c.canDowngradeVMProto.Store(true)
|
||||
// Auto-detect whether the remote storage supports VictoriaMetrics remote write protocol.
|
||||
doRequest := func(url string) (*http.Response, error) {
|
||||
return c.doRequest(url, nil)
|
||||
}
|
||||
useVMProto = protoparserutil.HandleVMProtoClientHandshake(c.remoteWriteURL, doRequest)
|
||||
if !useVMProto {
|
||||
logger.Infof("the remote storage at %q doesn't support VictoriaMetrics remote write protocol. Switching to Prometheus remote write protocol. "+
|
||||
"See https://docs.victoriametrics.com/vmagent/#victoriametrics-remote-write-protocol", sanitizedURL)
|
||||
}
|
||||
}
|
||||
c.useVMProto.Store(useVMProto)
|
||||
c.useVMProto = useVMProto
|
||||
|
||||
return c
|
||||
}
|
||||
@@ -382,7 +386,7 @@ func (c *client) newRequest(url string, body []byte) (*http.Request, error) {
|
||||
h := req.Header
|
||||
h.Set("User-Agent", "vmagent")
|
||||
h.Set("Content-Type", "application/x-protobuf")
|
||||
if encoding.IsZstd(body) {
|
||||
if c.useVMProto {
|
||||
h.Set("Content-Encoding", "zstd")
|
||||
h.Set("X-VictoriaMetrics-Remote-Write-Version", "1")
|
||||
} else {
|
||||
@@ -418,7 +422,7 @@ again:
|
||||
if retryDuration > maxRetryDuration {
|
||||
retryDuration = maxRetryDuration
|
||||
}
|
||||
remoteWriteRetryLogger.Warnf("couldn't send a block with size %d bytes to %q: %s; re-sending the block in %.3f seconds",
|
||||
logger.Warnf("couldn't send a block with size %d bytes to %q: %s; re-sending the block in %.3f seconds",
|
||||
len(block), c.sanitizedURL, err, retryDuration.Seconds())
|
||||
t := timerpool.Get(retryDuration)
|
||||
select {
|
||||
@@ -431,7 +435,6 @@ again:
|
||||
c.retriesCount.Inc()
|
||||
goto again
|
||||
}
|
||||
|
||||
statusCode := resp.StatusCode
|
||||
if statusCode/100 == 2 {
|
||||
_ = resp.Body.Close()
|
||||
@@ -440,43 +443,21 @@ again:
|
||||
c.blocksSent.Inc()
|
||||
return true
|
||||
}
|
||||
|
||||
metrics.GetOrCreateCounter(fmt.Sprintf(`vmagent_remotewrite_requests_total{url=%q, status_code="%d"}`, c.sanitizedURL, statusCode)).Inc()
|
||||
if statusCode == 409 {
|
||||
logBlockRejected(block, c.sanitizedURL, resp)
|
||||
|
||||
// Just drop block on 409 status code like Prometheus does.
|
||||
// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/873
|
||||
// and https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1149
|
||||
if statusCode == 409 || statusCode == 400 {
|
||||
body, err := io.ReadAll(resp.Body)
|
||||
_ = resp.Body.Close()
|
||||
c.packetsDropped.Inc()
|
||||
return true
|
||||
// - Remote Write v1 specification implicitly expects a `400 Bad Request` when the encoding is not supported.
|
||||
// - Remote Write v2 specification explicitly specifies a `415 Unsupported Media Type` for unsupported encodings.
|
||||
// - Real-world implementations of v1 use both 400 and 415 status codes.
|
||||
// See more in research: https://github.com/VictoriaMetrics/VictoriaMetrics/pull/8462#issuecomment-2786918054
|
||||
} else if statusCode == 415 || statusCode == 400 {
|
||||
if c.canDowngradeVMProto.Swap(false) {
|
||||
logger.Infof("received unsupported media type or bad request from remote storage at %q. Downgrading protocol from VictoriaMetrics to Prometheus remote write for all future requests. "+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/vmagent/#victoriametrics-remote-write-protocol", c.sanitizedURL)
|
||||
c.useVMProto.Store(false)
|
||||
if err != nil {
|
||||
remoteWriteRejectedLogger.Errorf("sending a block with size %d bytes to %q was rejected (skipping the block): status code %d; "+
|
||||
"failed to read response body: %s",
|
||||
len(block), c.sanitizedURL, statusCode, err)
|
||||
} else {
|
||||
remoteWriteRejectedLogger.Errorf("sending a block with size %d bytes to %q was rejected (skipping the block): status code %d; response body: %s",
|
||||
len(block), c.sanitizedURL, statusCode, string(body))
|
||||
}
|
||||
|
||||
if encoding.IsZstd(block) {
|
||||
logger.Infof("received unsupported media type or bad request from remote storage at %q. Re-packing the block to Prometheus remote write and retrying."+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/vmagent/#victoriametrics-remote-write-protocol", c.sanitizedURL)
|
||||
|
||||
block = mustRepackBlockFromZstdToSnappy(block)
|
||||
|
||||
c.retriesCount.Inc()
|
||||
_ = resp.Body.Close()
|
||||
goto again
|
||||
}
|
||||
|
||||
// Just drop snappy blocks on 400 or 415 status codes like Prometheus does.
|
||||
// Just drop block on 409 and 400 status codes like Prometheus does.
|
||||
// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/873
|
||||
// and https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1149
|
||||
logBlockRejected(block, c.sanitizedURL, resp)
|
||||
_ = resp.Body.Close()
|
||||
c.packetsDropped.Inc()
|
||||
return true
|
||||
@@ -509,7 +490,6 @@ again:
|
||||
}
|
||||
|
||||
var remoteWriteRejectedLogger = logger.WithThrottler("remoteWriteRejected", 5*time.Second)
|
||||
var remoteWriteRetryLogger = logger.WithThrottler("remoteWriteRetry", 5*time.Second)
|
||||
|
||||
// getRetryDuration returns retry duration.
|
||||
// retryAfterDuration has the highest priority.
|
||||
@@ -532,28 +512,6 @@ func getRetryDuration(retryAfterDuration, retryDuration, maxRetryDuration time.D
|
||||
return retryDuration
|
||||
}
|
||||
|
||||
func mustRepackBlockFromZstdToSnappy(zstdBlock []byte) []byte {
|
||||
plainBlock := make([]byte, 0, len(zstdBlock)*2)
|
||||
plainBlock, err := zstd.Decompress(plainBlock, zstdBlock)
|
||||
if err != nil {
|
||||
logger.Panicf("FATAL: cannot re-pack block with size %d bytes from Zstd to Snappy: %s", len(zstdBlock), err)
|
||||
}
|
||||
|
||||
return snappy.Encode(nil, plainBlock)
|
||||
}
|
||||
|
||||
func logBlockRejected(block []byte, sanitizedURL string, resp *http.Response) {
|
||||
body, err := io.ReadAll(resp.Body)
|
||||
if err != nil {
|
||||
remoteWriteRejectedLogger.Errorf("sending a block with size %d bytes to %q was rejected (skipping the block): status code %d; "+
|
||||
"failed to read response body: %s",
|
||||
len(block), sanitizedURL, resp.StatusCode, err)
|
||||
} else {
|
||||
remoteWriteRejectedLogger.Errorf("sending a block with size %d bytes to %q was rejected (skipping the block): status code %d; response body: %s",
|
||||
len(block), sanitizedURL, resp.StatusCode, string(body))
|
||||
}
|
||||
}
|
||||
|
||||
// parseRetryAfterHeader parses `Retry-After` value retrieved from HTTP response header.
|
||||
// retryAfterString should be in either HTTP-date or a number of seconds.
|
||||
// It will return time.Duration(0) if `retryAfterString` does not follow RFC 7231.
|
||||
|
||||
@@ -5,9 +5,6 @@ import (
|
||||
"net/http"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/encoding"
|
||||
"github.com/golang/snappy"
|
||||
)
|
||||
|
||||
func TestCalculateRetryDuration(t *testing.T) {
|
||||
@@ -100,19 +97,3 @@ func helper(d time.Duration) time.Duration {
|
||||
|
||||
return d + dv
|
||||
}
|
||||
|
||||
func TestRepackBlockFromZstdToSnappy(t *testing.T) {
|
||||
expectedPlainBlock := []byte(`foobar`)
|
||||
|
||||
zstdBlock := encoding.CompressZSTDLevel(nil, expectedPlainBlock, 1)
|
||||
snappyBlock := mustRepackBlockFromZstdToSnappy(zstdBlock)
|
||||
|
||||
actualPlainBlock, err := snappy.Decode(nil, snappyBlock)
|
||||
if err != nil {
|
||||
t.Fatalf("unexpected error: %s", err)
|
||||
}
|
||||
|
||||
if string(actualPlainBlock) != string(expectedPlainBlock) {
|
||||
t.Fatalf("unexpected plain block; got %q; want %q", actualPlainBlock, expectedPlainBlock)
|
||||
}
|
||||
}
|
||||
|
||||
@@ -16,7 +16,6 @@ import (
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/persistentqueue"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/promrelabel"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/slicesutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/timeutil"
|
||||
"github.com/VictoriaMetrics/metrics"
|
||||
"github.com/golang/snappy"
|
||||
@@ -29,7 +28,7 @@ var (
|
||||
maxRowsPerBlock = flag.Int("remoteWrite.maxRowsPerBlock", 10000, "The maximum number of samples to send in each block to remote storage. Higher number may improve performance at the cost of the increased memory usage. See also -remoteWrite.maxBlockSize")
|
||||
vmProtoCompressLevel = flag.Int("remoteWrite.vmProtoCompressLevel", 0, "The compression level for VictoriaMetrics remote write protocol. "+
|
||||
"Higher values reduce network traffic at the cost of higher CPU usage. Negative values reduce CPU usage at the cost of increased network traffic. "+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/vmagent/#victoriametrics-remote-write-protocol")
|
||||
"See https://docs.victoriametrics.com/vmagent/#victoriametrics-remote-write-protocol")
|
||||
)
|
||||
|
||||
type pendingSeries struct {
|
||||
@@ -40,7 +39,7 @@ type pendingSeries struct {
|
||||
periodicFlusherWG sync.WaitGroup
|
||||
}
|
||||
|
||||
func newPendingSeries(fq *persistentqueue.FastQueue, isVMRemoteWrite *atomic.Bool, significantFigures, roundDigits int) *pendingSeries {
|
||||
func newPendingSeries(fq *persistentqueue.FastQueue, isVMRemoteWrite bool, significantFigures, roundDigits int) *pendingSeries {
|
||||
var ps pendingSeries
|
||||
ps.wr.fq = fq
|
||||
ps.wr.isVMRemoteWrite = isVMRemoteWrite
|
||||
@@ -100,7 +99,7 @@ type writeRequest struct {
|
||||
fq *persistentqueue.FastQueue
|
||||
|
||||
// Whether to encode the write request with VictoriaMetrics remote write protocol.
|
||||
isVMRemoteWrite *atomic.Bool
|
||||
isVMRemoteWrite bool
|
||||
|
||||
// How many significant figures must be left before sending the writeRequest to fq.
|
||||
significantFigures int
|
||||
@@ -138,7 +137,7 @@ func (wr *writeRequest) reset() {
|
||||
// This is needed in order to properly save in-memory data to persistent queue on graceful shutdown.
|
||||
func (wr *writeRequest) mustFlushOnStop() {
|
||||
wr.wr.Timeseries = wr.tss
|
||||
if !tryPushWriteRequest(&wr.wr, wr.mustWriteBlock, wr.isVMRemoteWrite.Load()) {
|
||||
if !tryPushWriteRequest(&wr.wr, wr.mustWriteBlock, wr.isVMRemoteWrite) {
|
||||
logger.Panicf("BUG: final flush must always return true")
|
||||
}
|
||||
wr.reset()
|
||||
@@ -152,7 +151,7 @@ func (wr *writeRequest) mustWriteBlock(block []byte) bool {
|
||||
func (wr *writeRequest) tryFlush() bool {
|
||||
wr.wr.Timeseries = wr.tss
|
||||
wr.lastFlushTime.Store(fasttime.UnixTimestamp())
|
||||
if !tryPushWriteRequest(&wr.wr, wr.fq.TryWriteBlock, wr.isVMRemoteWrite.Load()) {
|
||||
if !tryPushWriteRequest(&wr.wr, wr.fq.TryWriteBlock, wr.isVMRemoteWrite) {
|
||||
return false
|
||||
}
|
||||
wr.reset()
|
||||
@@ -198,43 +197,28 @@ func (wr *writeRequest) tryPush(src []prompbmarshal.TimeSeries) bool {
|
||||
}
|
||||
|
||||
func (wr *writeRequest) copyTimeSeries(dst, src *prompbmarshal.TimeSeries) {
|
||||
labelsSrc := src.Labels
|
||||
|
||||
// Pre-allocate memory for labels.
|
||||
labelsDst := wr.labels
|
||||
labelsLen := len(wr.labels)
|
||||
wr.labels = slicesutil.SetLength(wr.labels, labelsLen+len(labelsSrc))
|
||||
labelsDst := wr.labels[labelsLen:]
|
||||
samplesDst := wr.samples
|
||||
buf := wr.buf
|
||||
for i := range src.Labels {
|
||||
labelsDst = append(labelsDst, prompbmarshal.Label{})
|
||||
dstLabel := &labelsDst[len(labelsDst)-1]
|
||||
srcLabel := &src.Labels[i]
|
||||
|
||||
// Pre-allocate memory for byte slice needed for storing label names and values.
|
||||
neededBufLen := 0
|
||||
for i := range labelsSrc {
|
||||
label := &labelsSrc[i]
|
||||
neededBufLen += len(label.Name) + len(label.Value)
|
||||
}
|
||||
bufLen := len(wr.buf)
|
||||
wr.buf = slicesutil.SetLength(wr.buf, bufLen+neededBufLen)
|
||||
buf := wr.buf[:bufLen]
|
||||
|
||||
// Copy labels
|
||||
for i := range labelsSrc {
|
||||
dstLabel := &labelsDst[i]
|
||||
srcLabel := &labelsSrc[i]
|
||||
|
||||
bufLen := len(buf)
|
||||
buf = append(buf, srcLabel.Name...)
|
||||
dstLabel.Name = bytesutil.ToUnsafeString(buf[bufLen:])
|
||||
|
||||
bufLen = len(buf)
|
||||
dstLabel.Name = bytesutil.ToUnsafeString(buf[len(buf)-len(srcLabel.Name):])
|
||||
buf = append(buf, srcLabel.Value...)
|
||||
dstLabel.Value = bytesutil.ToUnsafeString(buf[bufLen:])
|
||||
dstLabel.Value = bytesutil.ToUnsafeString(buf[len(buf)-len(srcLabel.Value):])
|
||||
}
|
||||
wr.buf = buf
|
||||
dst.Labels = labelsDst
|
||||
dst.Labels = labelsDst[labelsLen:]
|
||||
|
||||
// Copy samples
|
||||
samplesLen := len(wr.samples)
|
||||
wr.samples = append(wr.samples, src.Samples...)
|
||||
dst.Samples = wr.samples[samplesLen:]
|
||||
samplesDst = append(samplesDst, src.Samples...)
|
||||
dst.Samples = samplesDst[len(samplesDst)-len(src.Samples):]
|
||||
|
||||
wr.samples = samplesDst
|
||||
wr.labels = labelsDst
|
||||
wr.buf = buf
|
||||
}
|
||||
|
||||
// marshalConcurrency limits the maximum number of concurrent workers, which marshal and compress WriteRequest.
|
||||
|
||||
@@ -7,13 +7,10 @@ import (
|
||||
"strings"
|
||||
"sync"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/fasttime"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/prompbmarshal"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/promrelabel"
|
||||
|
||||
"github.com/VictoriaMetrics/metrics"
|
||||
)
|
||||
|
||||
var (
|
||||
@@ -22,10 +19,10 @@ var (
|
||||
relabelConfigPathGlobal = flag.String("remoteWrite.relabelConfig", "", "Optional path to file with relabeling configs, which are applied "+
|
||||
"to all the metrics before sending them to -remoteWrite.url. See also -remoteWrite.urlRelabelConfig. "+
|
||||
"The path can point either to local file or to http url. "+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/relabeling/")
|
||||
"See https://docs.victoriametrics.com/vmagent/#relabeling")
|
||||
relabelConfigPaths = flagutil.NewArrayString("remoteWrite.urlRelabelConfig", "Optional path to relabel configs for the corresponding -remoteWrite.url. "+
|
||||
"See also -remoteWrite.relabelConfig. The path can point either to local file or to http url. "+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/relabeling/")
|
||||
"See https://docs.victoriametrics.com/vmagent/#relabeling")
|
||||
|
||||
usePromCompatibleNaming = flag.Bool("usePromCompatibleNaming", false, "Whether to replace characters unsupported by Prometheus with underscores "+
|
||||
"in the ingested metric names and label names. For example, foo.bar{a.b='c'} is transformed into foo_bar{a_b='c'} during data ingestion if this flag is set. "+
|
||||
@@ -34,59 +31,12 @@ var (
|
||||
|
||||
var labelsGlobal []prompbmarshal.Label
|
||||
|
||||
var (
|
||||
relabelConfigReloads *metrics.Counter
|
||||
relabelConfigReloadErrors *metrics.Counter
|
||||
relabelConfigSuccess *metrics.Gauge
|
||||
relabelConfigTimestamp *metrics.Counter
|
||||
)
|
||||
|
||||
func initRelabelMetrics() {
|
||||
relabelConfigReloads = metrics.NewCounter(`vmagent_relabel_config_reloads_total`)
|
||||
relabelConfigReloadErrors = metrics.NewCounter(`vmagent_relabel_config_reloads_errors_total`)
|
||||
relabelConfigSuccess = metrics.NewGauge(`vmagent_relabel_config_last_reload_successful`, nil)
|
||||
relabelConfigTimestamp = metrics.NewCounter(`vmagent_relabel_config_last_reload_success_timestamp_seconds`)
|
||||
}
|
||||
|
||||
// CheckRelabelConfigs checks -remoteWrite.relabelConfig and -remoteWrite.urlRelabelConfig.
|
||||
func CheckRelabelConfigs() error {
|
||||
_, err := loadRelabelConfigs()
|
||||
return err
|
||||
}
|
||||
|
||||
func initRelabelConfigs() {
|
||||
rcs, err := loadRelabelConfigs()
|
||||
if err != nil {
|
||||
logger.Fatalf("cannot initialize relabel configs: %s", err)
|
||||
}
|
||||
allRelabelConfigs.Store(rcs)
|
||||
if rcs.isSet() {
|
||||
initRelabelMetrics()
|
||||
relabelConfigSuccess.Set(1)
|
||||
relabelConfigTimestamp.Set(fasttime.UnixTimestamp())
|
||||
}
|
||||
}
|
||||
|
||||
func reloadRelabelConfigs() {
|
||||
rcs := allRelabelConfigs.Load()
|
||||
if !rcs.isSet() {
|
||||
return
|
||||
}
|
||||
relabelConfigReloads.Inc()
|
||||
logger.Infof("reloading relabel configs pointed by -remoteWrite.relabelConfig and -remoteWrite.urlRelabelConfig")
|
||||
rcs, err := loadRelabelConfigs()
|
||||
if err != nil {
|
||||
relabelConfigReloadErrors.Inc()
|
||||
relabelConfigSuccess.Set(0)
|
||||
logger.Errorf("cannot reload relabel configs; preserving the previous configs; error: %s", err)
|
||||
return
|
||||
}
|
||||
allRelabelConfigs.Store(rcs)
|
||||
relabelConfigSuccess.Set(1)
|
||||
relabelConfigTimestamp.Set(fasttime.UnixTimestamp())
|
||||
logger.Infof("successfully reloaded relabel configs")
|
||||
}
|
||||
|
||||
func loadRelabelConfigs() (*relabelConfigs, error) {
|
||||
var rcs relabelConfigs
|
||||
if *relabelConfigPathGlobal != "" {
|
||||
@@ -120,21 +70,6 @@ type relabelConfigs struct {
|
||||
perURL []*promrelabel.ParsedConfigs
|
||||
}
|
||||
|
||||
func (rcs *relabelConfigs) isSet() bool {
|
||||
if rcs == nil {
|
||||
return false
|
||||
}
|
||||
if rcs.global.Len() > 0 {
|
||||
return true
|
||||
}
|
||||
for _, pc := range rcs.perURL {
|
||||
if pc.Len() > 0 {
|
||||
return true
|
||||
}
|
||||
}
|
||||
return false
|
||||
}
|
||||
|
||||
// initLabelsGlobal must be called after parsing command-line flags.
|
||||
func initLabelsGlobal() {
|
||||
labelsGlobal = nil
|
||||
|
||||
@@ -15,7 +15,7 @@ import (
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bloomfilter"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/cgroup"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/consistenthash"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/fasttime"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/fs"
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
||||
@@ -39,21 +39,22 @@ var (
|
||||
"Pass multiple -remoteWrite.url options in order to replicate the collected data to multiple remote storage systems. "+
|
||||
"The data can be sharded among the configured remote storage systems if -remoteWrite.shardByURL flag is set")
|
||||
enableMultitenantHandlers = flag.Bool("enableMultitenantHandlers", false, "Whether to process incoming data via multitenant insert handlers according to "+
|
||||
"https://docs.victoriametrics.com/victoriametrics/cluster-victoriametrics/#url-format . By default incoming data is processed via single-node insert handlers "+
|
||||
"according to https://docs.victoriametrics.com/victoriametrics/single-server-victoriametrics/#how-to-import-time-series-data ."+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/vmagent/#multitenancy for details")
|
||||
"https://docs.victoriametrics.com/cluster-victoriametrics/#url-format . By default incoming data is processed via single-node insert handlers "+
|
||||
"according to https://docs.victoriametrics.com/#how-to-import-time-series-data ."+
|
||||
"See https://docs.victoriametrics.com/vmagent/#multitenancy for details")
|
||||
|
||||
shardByURL = flag.Bool("remoteWrite.shardByURL", false, "Whether to shard outgoing series across all the remote storage systems enumerated via -remoteWrite.url. "+
|
||||
"By default the data is replicated across all the -remoteWrite.url . See https://docs.victoriametrics.com/victoriametrics/vmagent/#sharding-among-remote-storages . "+
|
||||
shardByURL = flag.Bool("remoteWrite.shardByURL", false, "Whether to shard outgoing series across all the remote storage systems enumerated via -remoteWrite.url . "+
|
||||
"By default the data is replicated across all the -remoteWrite.url . See https://docs.victoriametrics.com/vmagent/#sharding-among-remote-storages . "+
|
||||
"See also -remoteWrite.shardByURLReplicas")
|
||||
shardByURLReplicas = flag.Int("remoteWrite.shardByURLReplicas", 1, "How many copies of data to make among remote storage systems enumerated via -remoteWrite.url "+
|
||||
"when -remoteWrite.shardByURL is set. See https://docs.victoriametrics.com/victoriametrics/vmagent/#sharding-among-remote-storages")
|
||||
"when -remoteWrite.shardByURL is set. See https://docs.victoriametrics.com/vmagent/#sharding-among-remote-storages")
|
||||
shardByURLLabels = flagutil.NewArrayString("remoteWrite.shardByURL.labels", "Optional list of labels, which must be used for sharding outgoing samples "+
|
||||
"among remote storage systems if -remoteWrite.shardByURL command-line flag is set. By default all the labels are used for sharding in order to gain "+
|
||||
"even distribution of series over the specified -remoteWrite.url systems. See also -remoteWrite.shardByURL.ignoreLabels")
|
||||
shardByURLIgnoreLabels = flagutil.NewArrayString("remoteWrite.shardByURL.ignoreLabels", "Optional list of labels, which must be ignored when sharding outgoing samples "+
|
||||
"among remote storage systems if -remoteWrite.shardByURL command-line flag is set. By default all the labels are used for sharding in order to gain "+
|
||||
"even distribution of series over the specified -remoteWrite.url systems. See also -remoteWrite.shardByURL.labels")
|
||||
|
||||
tmpDataPath = flag.String("remoteWrite.tmpDataPath", "vmagent-remotewrite-data", "Path to directory for storing pending data, which isn't sent to the configured -remoteWrite.url . "+
|
||||
"See also -remoteWrite.maxDiskUsagePerURL and -remoteWrite.disableOnDiskQueue")
|
||||
keepDanglingQueues = flag.Bool("remoteWrite.keepDanglingQueues", false, "Keep persistent queues contents at -remoteWrite.tmpDataPath in case there are no matching -remoteWrite.url. "+
|
||||
@@ -80,30 +81,28 @@ var (
|
||||
`For example, if m{k1="v1",k2="v2"} may be sent as m{k2="v2",k1="v1"}`+
|
||||
`Enabled sorting for labels can slow down ingestion performance a bit`)
|
||||
maxHourlySeries = flag.Int("remoteWrite.maxHourlySeries", 0, "The maximum number of unique series vmagent can send to remote storage systems during the last hour. "+
|
||||
"Excess series are logged and dropped. This can be useful for limiting series cardinality. See https://docs.victoriametrics.com/victoriametrics/vmagent/#cardinality-limiter")
|
||||
"Excess series are logged and dropped. This can be useful for limiting series cardinality. See https://docs.victoriametrics.com/vmagent/#cardinality-limiter")
|
||||
maxDailySeries = flag.Int("remoteWrite.maxDailySeries", 0, "The maximum number of unique series vmagent can send to remote storage systems during the last 24 hours. "+
|
||||
"Excess series are logged and dropped. This can be useful for limiting series churn rate. See https://docs.victoriametrics.com/victoriametrics/vmagent/#cardinality-limiter")
|
||||
"Excess series are logged and dropped. This can be useful for limiting series churn rate. See https://docs.victoriametrics.com/vmagent/#cardinality-limiter")
|
||||
maxIngestionRate = flag.Int("maxIngestionRate", 0, "The maximum number of samples vmagent can receive per second. Data ingestion is paused when the limit is exceeded. "+
|
||||
"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/victoriametrics/vmagent/#disabling-on-disk-persistence . See also -remoteWrite.dropSamplesOnOverload")
|
||||
"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/victoriametrics/vmagent/#disabling-on-disk-persistence")
|
||||
"cannot be pushed into the configured -remoteWrite.url 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
|
||||
rwctxsGlobalIdx []int
|
||||
rwctxConsistentHashGlobal *consistenthash.ConsistentHash
|
||||
rwctxsGlobal []*remoteWriteCtx
|
||||
|
||||
// ErrQueueFullHTTPRetry must be returned when TryPush() returns false.
|
||||
ErrQueueFullHTTPRetry = &httpserver.ErrorWithStatusCode{
|
||||
Err: fmt.Errorf("remote storage systems cannot keep up with the data ingestion rate; retry the request later " +
|
||||
"or remove -remoteWrite.disableOnDiskQueue from vmagent command-line flags, so it could save pending data to -remoteWrite.tmpDataPath; " +
|
||||
"see https://docs.victoriametrics.com/victoriametrics/vmagent/#disabling-on-disk-persistence"),
|
||||
"see https://docs.victoriametrics.com/vmagent/#disabling-on-disk-persistence"),
|
||||
StatusCode: http.StatusTooManyRequests,
|
||||
}
|
||||
|
||||
@@ -184,7 +183,7 @@ func Init() {
|
||||
|
||||
if len(*shardByURLLabels) > 0 && len(*shardByURLIgnoreLabels) > 0 {
|
||||
logger.Fatalf("-remoteWrite.shardByURL.labels and -remoteWrite.shardByURL.ignoreLabels cannot be set simultaneously; " +
|
||||
"see https://docs.victoriametrics.com/victoriametrics/vmagent/#sharding-among-remote-storages")
|
||||
"see https://docs.victoriametrics.com/vmagent/#sharding-among-remote-storages")
|
||||
}
|
||||
shardByURLLabelsMap = newMapFromStrings(*shardByURLLabels)
|
||||
shardByURLIgnoreLabelsMap = newMapFromStrings(*shardByURLIgnoreLabels)
|
||||
@@ -196,11 +195,17 @@ func Init() {
|
||||
// See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1240
|
||||
sighupCh := procutil.NewSighupChan()
|
||||
|
||||
initRelabelConfigs()
|
||||
rcs, err := loadRelabelConfigs()
|
||||
if err != nil {
|
||||
logger.Fatalf("cannot load relabel configs: %s", err)
|
||||
}
|
||||
allRelabelConfigs.Store(rcs)
|
||||
relabelConfigSuccess.Set(1)
|
||||
relabelConfigTimestamp.Set(fasttime.UnixTimestamp())
|
||||
|
||||
initStreamAggrConfigGlobal()
|
||||
|
||||
initRemoteWriteCtxs(*remoteWriteURLs)
|
||||
rwctxsGlobal = newRemoteWriteCtxs(*remoteWriteURLs)
|
||||
|
||||
disableOnDiskQueues := []bool(*disableOnDiskQueue)
|
||||
disableOnDiskQueueAny = slices.Contains(disableOnDiskQueues, true)
|
||||
@@ -262,7 +267,30 @@ func dropDanglingQueues() {
|
||||
}
|
||||
}
|
||||
|
||||
func initRemoteWriteCtxs(urls []string) {
|
||||
func reloadRelabelConfigs() {
|
||||
relabelConfigReloads.Inc()
|
||||
logger.Infof("reloading relabel configs pointed by -remoteWrite.relabelConfig and -remoteWrite.urlRelabelConfig")
|
||||
rcs, err := loadRelabelConfigs()
|
||||
if err != nil {
|
||||
relabelConfigReloadErrors.Inc()
|
||||
relabelConfigSuccess.Set(0)
|
||||
logger.Errorf("cannot reload relabel configs; preserving the previous configs; error: %s", err)
|
||||
return
|
||||
}
|
||||
allRelabelConfigs.Store(rcs)
|
||||
relabelConfigSuccess.Set(1)
|
||||
relabelConfigTimestamp.Set(fasttime.UnixTimestamp())
|
||||
logger.Infof("successfully reloaded relabel configs")
|
||||
}
|
||||
|
||||
var (
|
||||
relabelConfigReloads = metrics.NewCounter(`vmagent_relabel_config_reloads_total`)
|
||||
relabelConfigReloadErrors = metrics.NewCounter(`vmagent_relabel_config_reloads_errors_total`)
|
||||
relabelConfigSuccess = metrics.NewGauge(`vmagent_relabel_config_last_reload_successful`, nil)
|
||||
relabelConfigTimestamp = metrics.NewCounter(`vmagent_relabel_config_last_reload_success_timestamp_seconds`)
|
||||
)
|
||||
|
||||
func newRemoteWriteCtxs(urls []string) []*remoteWriteCtx {
|
||||
if len(urls) == 0 {
|
||||
logger.Panicf("BUG: urls must be non-empty")
|
||||
}
|
||||
@@ -278,7 +306,6 @@ func initRemoteWriteCtxs(urls []string) {
|
||||
maxInmemoryBlocks = 2
|
||||
}
|
||||
rwctxs := make([]*remoteWriteCtx, len(urls))
|
||||
rwctxIdx := make([]int, len(urls))
|
||||
for i, remoteWriteURLRaw := range urls {
|
||||
remoteWriteURL, err := url.Parse(remoteWriteURLRaw)
|
||||
if err != nil {
|
||||
@@ -289,19 +316,8 @@ func initRemoteWriteCtxs(urls []string) {
|
||||
sanitizedURL = fmt.Sprintf("%d:%s", i+1, remoteWriteURL)
|
||||
}
|
||||
rwctxs[i] = newRemoteWriteCtx(i, remoteWriteURL, maxInmemoryBlocks, sanitizedURL)
|
||||
rwctxIdx[i] = i
|
||||
}
|
||||
|
||||
if *shardByURL {
|
||||
consistentHashNodes := make([]string, 0, len(urls))
|
||||
for i, url := range urls {
|
||||
consistentHashNodes = append(consistentHashNodes, fmt.Sprintf("%d:%s", i+1, url))
|
||||
}
|
||||
rwctxConsistentHashGlobal = consistenthash.NewConsistentHash(consistentHashNodes, 0)
|
||||
}
|
||||
|
||||
rwctxsGlobal = rwctxs
|
||||
rwctxsGlobalIdx = rwctxIdx
|
||||
return rwctxs
|
||||
}
|
||||
|
||||
var (
|
||||
@@ -485,10 +501,6 @@ func tryPush(at *auth.Token, wr *prompbmarshal.WriteRequest, forceDropSamplesOnF
|
||||
return true
|
||||
}
|
||||
|
||||
// getEligibleRemoteWriteCtxs checks whether writes to configured remote storage systems are blocked and
|
||||
// returns only the unblocked rwctx.
|
||||
//
|
||||
// calculateHealthyRwctxIdx will rely on the order of rwctx to be in ascending order.
|
||||
func getEligibleRemoteWriteCtxs(tss []prompbmarshal.TimeSeries, forceDropSamplesOnFailure bool) ([]*remoteWriteCtx, bool) {
|
||||
if !disableOnDiskQueueAny {
|
||||
return rwctxsGlobal, true
|
||||
@@ -505,12 +517,6 @@ func getEligibleRemoteWriteCtxs(tss []prompbmarshal.TimeSeries, forceDropSamples
|
||||
return nil, false
|
||||
}
|
||||
rowsCount := getRowsCount(tss)
|
||||
if *shardByURL {
|
||||
// Todo: When shardByURL is enabled, the following metrics won't be 100% accurate. Because vmagent don't know
|
||||
// which rwctx should data be pushed to yet. Let's consider the hashing algorithm fair and will distribute
|
||||
// data to all rwctxs evenly.
|
||||
rowsCount = rowsCount / len(rwctxsGlobal)
|
||||
}
|
||||
rwctx.rowsDroppedOnPushFailure.Add(rowsCount)
|
||||
}
|
||||
}
|
||||
@@ -522,7 +528,6 @@ func pushToRemoteStoragesTrackDropped(tss []prompbmarshal.TimeSeries) {
|
||||
if len(rwctxs) == 0 {
|
||||
return
|
||||
}
|
||||
|
||||
if !tryPushBlockToRemoteStorages(rwctxs, tss, true) {
|
||||
logger.Panicf("BUG: tryPushBlockToRemoteStorages() must return true when forceDropSamplesOnFailure=true")
|
||||
}
|
||||
@@ -573,68 +578,7 @@ func tryShardingBlockAmongRemoteStorages(rwctxs []*remoteWriteCtx, tssBlock []pr
|
||||
defer putTSSShards(x)
|
||||
|
||||
shards := x.shards
|
||||
shardAmountRemoteWriteCtx(tssBlock, shards, rwctxs, replicas)
|
||||
|
||||
// Push sharded samples to remote storage systems in parallel in order to reduce
|
||||
// the time needed for sending the data to multiple remote storage systems.
|
||||
var wg sync.WaitGroup
|
||||
var anyPushFailed atomic.Bool
|
||||
for i, rwctx := range rwctxs {
|
||||
shard := shards[i]
|
||||
if len(shard) == 0 {
|
||||
continue
|
||||
}
|
||||
wg.Add(1)
|
||||
go func(rwctx *remoteWriteCtx, tss []prompbmarshal.TimeSeries) {
|
||||
defer wg.Done()
|
||||
if !rwctx.TryPush(tss, forceDropSamplesOnFailure) {
|
||||
anyPushFailed.Store(true)
|
||||
}
|
||||
}(rwctx, shard)
|
||||
}
|
||||
wg.Wait()
|
||||
return !anyPushFailed.Load()
|
||||
}
|
||||
|
||||
// calculateHealthyRwctxIdx returns the index of healthyRwctxs in rwctxsGlobal.
|
||||
// It relies on the order of rwctx in healthyRwctxs, which is appended by getEligibleRemoteWriteCtxs.
|
||||
func calculateHealthyRwctxIdx(healthyRwctxs []*remoteWriteCtx) ([]int, []int) {
|
||||
// fast path: all rwctxs are healthy.
|
||||
if len(healthyRwctxs) == len(rwctxsGlobal) {
|
||||
return rwctxsGlobalIdx, nil
|
||||
}
|
||||
|
||||
unhealthyIdx := make([]int, 0, len(rwctxsGlobal))
|
||||
healthyIdx := make([]int, 0, len(rwctxsGlobal))
|
||||
|
||||
var i int
|
||||
for j := range rwctxsGlobal {
|
||||
if i < len(healthyRwctxs) && rwctxsGlobal[j].idx == healthyRwctxs[i].idx {
|
||||
healthyIdx = append(healthyIdx, j)
|
||||
i++
|
||||
} else {
|
||||
unhealthyIdx = append(unhealthyIdx, j)
|
||||
}
|
||||
}
|
||||
|
||||
return healthyIdx, unhealthyIdx
|
||||
}
|
||||
|
||||
// shardAmountRemoteWriteCtx distribute time series to shards by consistent hashing.
|
||||
func shardAmountRemoteWriteCtx(tssBlock []prompbmarshal.TimeSeries, shards [][]prompbmarshal.TimeSeries, rwctxs []*remoteWriteCtx, replicas int) {
|
||||
tmpLabels := promutil.GetLabels()
|
||||
defer promutil.PutLabels(tmpLabels)
|
||||
|
||||
healthyIdx, unhealthyIdx := calculateHealthyRwctxIdx(rwctxs)
|
||||
|
||||
// shardsIdxMap is a map to find which the shard idx by rwctxs idx.
|
||||
// rwctxConsistentHashGlobal will tell which the rwctxs idx a time series should be written to.
|
||||
// And this time series should be appended to the shards by correct shard idx.
|
||||
shardsIdxMap := make(map[int]int, len(healthyIdx))
|
||||
for idx, rwctxsIdx := range healthyIdx {
|
||||
shardsIdxMap[rwctxsIdx] = idx
|
||||
}
|
||||
|
||||
for _, ts := range tssBlock {
|
||||
hashLabels := ts.Labels
|
||||
if len(shardByURLLabelsMap) > 0 {
|
||||
@@ -655,25 +599,41 @@ func shardAmountRemoteWriteCtx(tssBlock []prompbmarshal.TimeSeries, shards [][]p
|
||||
tmpLabels.Labels = hashLabels
|
||||
}
|
||||
h := getLabelsHash(hashLabels)
|
||||
|
||||
// Get the rwctxIdx through consistent hashing and then map it to the index in shards.
|
||||
// The rwctxIdx is not always equal to the shardIdx, for example, when some rwctx are not available.
|
||||
rwctxIdx := rwctxConsistentHashGlobal.GetNodeIdx(h, unhealthyIdx)
|
||||
shardIdx := shardsIdxMap[rwctxIdx]
|
||||
|
||||
replicated := 0
|
||||
idx := h % uint64(len(shards))
|
||||
i := 0
|
||||
for {
|
||||
shards[shardIdx] = append(shards[shardIdx], ts)
|
||||
replicated++
|
||||
if replicated >= replicas {
|
||||
shards[idx] = append(shards[idx], ts)
|
||||
i++
|
||||
if i >= replicas {
|
||||
break
|
||||
}
|
||||
shardIdx++
|
||||
if shardIdx >= len(shards) {
|
||||
shardIdx = 0
|
||||
idx++
|
||||
if idx >= uint64(len(shards)) {
|
||||
idx = 0
|
||||
}
|
||||
}
|
||||
}
|
||||
promutil.PutLabels(tmpLabels)
|
||||
|
||||
// Push sharded samples to remote storage systems in parallel in order to reduce
|
||||
// the time needed for sending the data to multiple remote storage systems.
|
||||
var wg sync.WaitGroup
|
||||
var anyPushFailed atomic.Bool
|
||||
for i, rwctx := range rwctxs {
|
||||
shard := shards[i]
|
||||
if len(shard) == 0 {
|
||||
continue
|
||||
}
|
||||
wg.Add(1)
|
||||
go func(rwctx *remoteWriteCtx, tss []prompbmarshal.TimeSeries) {
|
||||
defer wg.Done()
|
||||
if !rwctx.TryPush(tss, forceDropSamplesOnFailure) {
|
||||
anyPushFailed.Store(true)
|
||||
}
|
||||
}(rwctx, shard)
|
||||
}
|
||||
wg.Wait()
|
||||
return !anyPushFailed.Load()
|
||||
}
|
||||
|
||||
type tssShards struct {
|
||||
@@ -847,7 +807,7 @@ func newRemoteWriteCtx(argIdx int, remoteWriteURL *url.URL, maxInmemoryBlocks in
|
||||
}
|
||||
pss := make([]*pendingSeries, pssLen)
|
||||
for i := range pss {
|
||||
pss[i] = newPendingSeries(fq, &c.useVMProto, sf, rd)
|
||||
pss[i] = newPendingSeries(fq, c.useVMProto, sf, rd)
|
||||
}
|
||||
|
||||
rwctx := &remoteWriteCtx{
|
||||
|
||||
@@ -4,17 +4,13 @@ import (
|
||||
"fmt"
|
||||
"math"
|
||||
"reflect"
|
||||
"strconv"
|
||||
"sync/atomic"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/VictoriaMetrics/VictoriaMetrics/lib/consistenthash"
|
||||
"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"
|
||||
)
|
||||
|
||||
@@ -72,9 +68,7 @@ func TestRemoteWriteContext_TryPush_ImmutableTimeseries(t *testing.T) {
|
||||
allRelabelConfigs.Store(rcs)
|
||||
|
||||
pss := make([]*pendingSeries, 1)
|
||||
isVMProto := &atomic.Bool{}
|
||||
isVMProto.Store(true)
|
||||
pss[0] = newPendingSeries(nil, isVMProto, 0, 100)
|
||||
pss[0] = newPendingSeries(nil, true, 0, 100)
|
||||
rwctx := &remoteWriteCtx{
|
||||
idx: 0,
|
||||
streamAggrKeepInput: keepInput,
|
||||
@@ -177,173 +171,3 @@ metric{env="dev"} 15
|
||||
metric{env="bar"} 25
|
||||
`)
|
||||
}
|
||||
|
||||
func TestShardAmountRemoteWriteCtx(t *testing.T) {
|
||||
// 1. distribute 100000 series to n nodes.
|
||||
// 2. remove the last node from healthy list.
|
||||
// 3. distribute the same 10000 series to (n-1) node again.
|
||||
// 4. check active time series change rate:
|
||||
// change rate must < (3/total nodes). e.g. +30% if 10 you have 10 nodes.
|
||||
|
||||
f := func(remoteWriteCount int, healthyIdx []int, replicas int) {
|
||||
t.Helper()
|
||||
defer func() {
|
||||
rwctxsGlobal = nil
|
||||
rwctxsGlobalIdx = nil
|
||||
rwctxConsistentHashGlobal = nil
|
||||
}()
|
||||
|
||||
rwctxsGlobal = make([]*remoteWriteCtx, remoteWriteCount)
|
||||
rwctxsGlobalIdx = make([]int, remoteWriteCount)
|
||||
rwctxs := make([]*remoteWriteCtx, 0, len(healthyIdx))
|
||||
|
||||
for i := range remoteWriteCount {
|
||||
rwCtx := &remoteWriteCtx{
|
||||
idx: i,
|
||||
}
|
||||
rwctxsGlobalIdx[i] = i
|
||||
|
||||
if i >= len(healthyIdx) {
|
||||
rwctxsGlobal[i] = rwCtx
|
||||
continue
|
||||
}
|
||||
hIdx := healthyIdx[i]
|
||||
if hIdx != i {
|
||||
rwctxs = append(rwctxs, &remoteWriteCtx{
|
||||
idx: hIdx,
|
||||
})
|
||||
} else {
|
||||
rwctxs = append(rwctxs, rwCtx)
|
||||
}
|
||||
rwctxsGlobal[i] = rwCtx
|
||||
}
|
||||
|
||||
seriesCount := 100000
|
||||
// build 1000000 series
|
||||
tssBlock := make([]prompbmarshal.TimeSeries, 0, seriesCount)
|
||||
for i := 0; i < seriesCount; i++ {
|
||||
tssBlock = append(tssBlock, prompbmarshal.TimeSeries{
|
||||
Labels: []prompbmarshal.Label{
|
||||
{
|
||||
Name: "label",
|
||||
Value: strconv.Itoa(i),
|
||||
},
|
||||
},
|
||||
Samples: []prompbmarshal.Sample{
|
||||
{
|
||||
Timestamp: 0,
|
||||
Value: 0,
|
||||
},
|
||||
},
|
||||
})
|
||||
}
|
||||
|
||||
// build consistent hash for x remote write context
|
||||
// build active time series set
|
||||
nodes := make([]string, 0, remoteWriteCount)
|
||||
activeTimeSeriesByNodes := make([]map[string]struct{}, remoteWriteCount)
|
||||
for i := 0; i < remoteWriteCount; i++ {
|
||||
nodes = append(nodes, fmt.Sprintf("node%d", i))
|
||||
activeTimeSeriesByNodes[i] = make(map[string]struct{})
|
||||
}
|
||||
rwctxConsistentHashGlobal = consistenthash.NewConsistentHash(nodes, 0)
|
||||
|
||||
// create shards
|
||||
x := getTSSShards(len(rwctxs))
|
||||
shards := x.shards
|
||||
|
||||
// execute
|
||||
shardAmountRemoteWriteCtx(tssBlock, shards, rwctxs, replicas)
|
||||
|
||||
for i, nodeIdx := range healthyIdx {
|
||||
for _, ts := range shards[i] {
|
||||
// add it to node[nodeIdx]'s active time series
|
||||
activeTimeSeriesByNodes[nodeIdx][prompbmarshal.LabelsToString(ts.Labels)] = struct{}{}
|
||||
}
|
||||
}
|
||||
|
||||
totalActiveTimeSeries := 0
|
||||
for _, activeTimeSeries := range activeTimeSeriesByNodes {
|
||||
totalActiveTimeSeries += len(activeTimeSeries)
|
||||
}
|
||||
avgActiveTimeSeries1 := totalActiveTimeSeries / remoteWriteCount
|
||||
putTSSShards(x)
|
||||
|
||||
// removed last node
|
||||
rwctxs = rwctxs[:len(rwctxs)-1]
|
||||
healthyIdx = healthyIdx[:len(healthyIdx)-1]
|
||||
|
||||
x = getTSSShards(len(rwctxs))
|
||||
shards = x.shards
|
||||
|
||||
// execute
|
||||
shardAmountRemoteWriteCtx(tssBlock, shards, rwctxs, replicas)
|
||||
for i, nodeIdx := range healthyIdx {
|
||||
for _, ts := range shards[i] {
|
||||
// add it to node[nodeIdx]'s active time series
|
||||
activeTimeSeriesByNodes[nodeIdx][prompbmarshal.LabelsToString(ts.Labels)] = struct{}{}
|
||||
}
|
||||
}
|
||||
|
||||
totalActiveTimeSeries = 0
|
||||
for _, activeTimeSeries := range activeTimeSeriesByNodes {
|
||||
totalActiveTimeSeries += len(activeTimeSeries)
|
||||
}
|
||||
avgActiveTimeSeries2 := totalActiveTimeSeries / remoteWriteCount
|
||||
|
||||
changed := math.Abs(float64(avgActiveTimeSeries2-avgActiveTimeSeries1) / float64(avgActiveTimeSeries1))
|
||||
threshold := 3 / float64(remoteWriteCount)
|
||||
|
||||
if changed >= threshold {
|
||||
t.Fatalf("average active time series before: %d, after: %d, changed: %.2f. threshold: %.2f", avgActiveTimeSeries1, avgActiveTimeSeries2, changed, threshold)
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
f(5, []int{0, 1, 2, 3, 4}, 1)
|
||||
|
||||
f(5, []int{0, 1, 2, 3, 4}, 2)
|
||||
|
||||
f(10, []int{0, 1, 2, 3, 4, 5, 6, 7, 9}, 1)
|
||||
|
||||
f(10, []int{0, 1, 2, 3, 4, 5, 6, 7, 9}, 3)
|
||||
}
|
||||
|
||||
func TestCalculateHealthyRwctxIdx(t *testing.T) {
|
||||
f := func(total int, healthyIdx []int, unhealthyIdx []int) {
|
||||
t.Helper()
|
||||
|
||||
healthyMap := make(map[int]bool)
|
||||
for _, idx := range healthyIdx {
|
||||
healthyMap[idx] = true
|
||||
}
|
||||
rwctxsGlobal = make([]*remoteWriteCtx, total)
|
||||
rwctxsGlobalIdx = make([]int, total)
|
||||
rwctxs := make([]*remoteWriteCtx, 0, len(healthyIdx))
|
||||
for i := range rwctxsGlobal {
|
||||
rwctx := &remoteWriteCtx{idx: i}
|
||||
rwctxsGlobal[i] = rwctx
|
||||
if healthyMap[i] {
|
||||
rwctxs = append(rwctxs, rwctx)
|
||||
}
|
||||
rwctxsGlobalIdx[i] = i
|
||||
}
|
||||
|
||||
gotHealthyIdx, gotUnhealthyIdx := calculateHealthyRwctxIdx(rwctxs)
|
||||
if !reflect.DeepEqual(healthyIdx, gotHealthyIdx) {
|
||||
t.Errorf("calculateHealthyRwctxIdx want healthyIdx = %v, got %v", healthyIdx, gotHealthyIdx)
|
||||
}
|
||||
if !reflect.DeepEqual(unhealthyIdx, gotUnhealthyIdx) {
|
||||
t.Errorf("calculateHealthyRwctxIdx want unhealthyIdx = %v, got %v", unhealthyIdx, gotUnhealthyIdx)
|
||||
}
|
||||
}
|
||||
|
||||
f(5, []int{0, 1, 2, 3, 4}, nil)
|
||||
f(5, []int{0, 1, 2, 4}, []int{3})
|
||||
f(5, []int{2, 4}, []int{0, 1, 3})
|
||||
f(5, []int{0, 2, 4}, []int{1, 3})
|
||||
f(5, []int{}, []int{0, 1, 2, 3, 4})
|
||||
f(5, []int{4}, []int{0, 1, 2, 3})
|
||||
f(1, []int{0}, nil)
|
||||
f(1, []int{}, []int{0})
|
||||
}
|
||||
|
||||
@@ -16,55 +16,55 @@ import (
|
||||
var (
|
||||
// Global config
|
||||
streamAggrGlobalConfig = flag.String("streamAggr.config", "", "Optional path to file with stream aggregation config. "+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/ . "+
|
||||
"See https://docs.victoriametrics.com/stream-aggregation/ . "+
|
||||
"See also -streamAggr.keepInput, -streamAggr.dropInput and -streamAggr.dedupInterval")
|
||||
streamAggrGlobalKeepInput = flag.Bool("streamAggr.keepInput", false, "Whether to keep all the input samples after the aggregation "+
|
||||
"with -streamAggr.config. By default, only aggregates samples are dropped, while the remaining samples "+
|
||||
"are written to remote storages write. See also -streamAggr.dropInput and https://docs.victoriametrics.com/victoriametrics/stream-aggregation/")
|
||||
"are written to remote storages write. See also -streamAggr.dropInput and https://docs.victoriametrics.com/stream-aggregation/")
|
||||
streamAggrGlobalDropInput = flag.Bool("streamAggr.dropInput", false, "Whether to drop all the input samples after the aggregation "+
|
||||
"with -remoteWrite.streamAggr.config. By default, only aggregates samples are dropped, while the remaining samples "+
|
||||
"are written to remote storages write. See also -streamAggr.keepInput and https://docs.victoriametrics.com/victoriametrics/stream-aggregation/")
|
||||
"are written to remote storages write. See also -streamAggr.keepInput and https://docs.victoriametrics.com/stream-aggregation/")
|
||||
streamAggrGlobalDedupInterval = flag.Duration("streamAggr.dedupInterval", 0, "Input samples are de-duplicated with this interval on "+
|
||||
"aggregator before optional aggregation with -streamAggr.config . "+
|
||||
"See also -dedup.minScrapeInterval and https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#deduplication")
|
||||
"See also -dedup.minScrapeInterval and https://docs.victoriametrics.com/stream-aggregation/#deduplication")
|
||||
streamAggrGlobalIgnoreOldSamples = flag.Bool("streamAggr.ignoreOldSamples", false, "Whether to ignore input samples with old timestamps outside the "+
|
||||
"current aggregation interval for aggregator. "+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#ignoring-old-samples")
|
||||
"See https://docs.victoriametrics.com/stream-aggregation/#ignoring-old-samples")
|
||||
streamAggrGlobalIgnoreFirstIntervals = flag.Int("streamAggr.ignoreFirstIntervals", 0, "Number of aggregation intervals to skip after the start for "+
|
||||
"aggregator. Increase this value if you observe incorrect aggregation results after vmagent restarts. It could be caused by receiving unordered delayed data from "+
|
||||
"clients pushing data into the vmagent. See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#ignore-aggregation-intervals-on-start")
|
||||
"clients pushing data into the vmagent. See https://docs.victoriametrics.com/stream-aggregation/#ignore-aggregation-intervals-on-start")
|
||||
streamAggrGlobalDropInputLabels = flagutil.NewArrayString("streamAggr.dropInputLabels", "An optional list of labels to drop from samples for aggregator "+
|
||||
"before stream de-duplication and aggregation . See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#dropping-unneeded-labels")
|
||||
"before stream de-duplication and aggregation . See https://docs.victoriametrics.com/stream-aggregation/#dropping-unneeded-labels")
|
||||
streamAggrGlobalEnableWindows = flag.Bool("streamAggr.enableWindows", false, "Enables aggregation within fixed windows for all global aggregators. "+
|
||||
"This allows to get more precise results, but impacts resource usage as it requires twice more memory to store two states. "+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#aggregation-windows.")
|
||||
"See https://docs.victoriametrics.com/stream-aggregation/#aggregation-windows.")
|
||||
|
||||
// Per URL config
|
||||
streamAggrConfig = flagutil.NewArrayString("remoteWrite.streamAggr.config", "Optional path to file with stream aggregation config for the corresponding -remoteWrite.url. "+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/ . "+
|
||||
"See 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 "+
|
||||
"are written to the corresponding -remoteWrite.url . See also -remoteWrite.streamAggr.keepInput and https://docs.victoriametrics.com/victoriametrics/stream-aggregation/")
|
||||
"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 "+
|
||||
"are written to the corresponding -remoteWrite.url . See also -remoteWrite.streamAggr.dropInput and https://docs.victoriametrics.com/victoriametrics/stream-aggregation/")
|
||||
"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/victoriametrics/stream-aggregation/#deduplication")
|
||||
"with -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. See also -dedup.minScrapeInterval and https://docs.victoriametrics.com/stream-aggregation/#deduplication")
|
||||
streamAggrIgnoreOldSamples = flagutil.NewArrayBool("remoteWrite.streamAggr.ignoreOldSamples", "Whether to ignore input samples with old timestamps outside the current "+
|
||||
"aggregation interval for the corresponding -remoteWrite.streamAggr.config at the corresponding -remoteWrite.url. "+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#ignoring-old-samples")
|
||||
"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 buffered delayed data from clients pushing data into the vmagent. "+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#ignore-aggregation-intervals-on-start")
|
||||
"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. "+
|
||||
"Multiple labels per remoteWrite.url must be delimited by '^^': -remoteWrite.streamAggr.dropInputLabels='replica^^az,replica'. "+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#dropping-unneeded-labels")
|
||||
"See https://docs.victoriametrics.com/stream-aggregation/#dropping-unneeded-labels")
|
||||
streamAggrEnableWindows = flagutil.NewArrayBool("remoteWrite.streamAggr.enableWindows", "Enables aggregation within fixed windows for all remote write's aggregators. "+
|
||||
"This allows to get more precise results, but impacts resource usage as it requires twice more memory to store two states. "+
|
||||
"See https://docs.victoriametrics.com/victoriametrics/stream-aggregation/#aggregation-windows.")
|
||||
"See https://docs.victoriametrics.com/stream-aggregation/#aggregation-windows.")
|
||||
)
|
||||
|
||||
// CheckStreamAggrConfigs checks -remoteWrite.streamAggr.config and -streamAggr.config.
|
||||
|
||||
@@ -1,3 +1,3 @@
|
||||
See vmalert-tool docs [here](https://docs.victoriametrics.com/victoriametrics/vmalert-tool/).
|
||||
See vmalert-tool docs [here](https://docs.victoriametrics.com/vmalert-tool.html).
|
||||
|
||||
vmalert-tool docs can be edited at [docs/vmalert-tool.md](https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/docs/victoriametrics/vmalert-tool.md).
|
||||
vmalert-tool docs can be edited at [docs/vmalert-tool.md](https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/docs/vmalert-tool.md).
|
||||
|
||||
@@ -1,8 +0,0 @@
|
||||
ARG base_image=non-existing
|
||||
FROM $base_image
|
||||
|
||||
EXPOSE 8880
|
||||
|
||||
ENTRYPOINT ["/vmalert-tool-prod"]
|
||||
ARG src_binary=non-existing
|
||||
COPY $src_binary ./vmalert-tool-prod
|
||||
@@ -17,13 +17,13 @@ func main() {
|
||||
app := &cli.App{
|
||||
Name: "vmalert-tool",
|
||||
Usage: "VMAlert command-line tool",
|
||||
UsageText: "More info in https://docs.victoriametrics.com/victoriametrics/vmalert-tool/",
|
||||
UsageText: "More info in https://docs.victoriametrics.com/vmalert-tool.html",
|
||||
Version: buildinfo.Version,
|
||||
Commands: []*cli.Command{
|
||||
{
|
||||
Name: "unittest",
|
||||
Usage: "Run unittest for alerting and recording rules.",
|
||||
UsageText: "More info in https://docs.victoriametrics.com/victoriametrics/vmalert-tool/#unit-testing-for-rules",
|
||||
UsageText: "More info in https://docs.victoriametrics.com/vmalert-tool.html#Unit-testing-for-rules",
|
||||
Flags: []cli.Flag{
|
||||
&cli.StringSliceFlag{
|
||||
Name: "files",
|
||||
|
||||
@@ -9,5 +9,4 @@ COPY --from=certs /etc/ssl/certs/ca-certificates.crt /etc/ssl/certs/ca-certifica
|
||||
EXPOSE 8429
|
||||
ENTRYPOINT ["/vmalert-tool-prod"]
|
||||
ARG TARGETARCH
|
||||
ARG BINARY_SUFFIX=non-existing
|
||||
COPY vmalert-tool-linux-${TARGETARCH}-prod${BINARY_SUFFIX} ./vmalert-tool-prod
|
||||
COPY vmalert-tool-linux-${TARGETARCH}-prod ./vmalert-tool-prod
|
||||
|
||||
@@ -18,7 +18,7 @@ import (
|
||||
"github.com/VictoriaMetrics/metricsql"
|
||||
)
|
||||
|
||||
var numReg = regexp.MustCompile(`(?i)[+x-]?(?:\d+(?:\.\d*)?|\.\d+|inf|nan|_)(?:e[+-]?\d+)?[+x-]?`)
|
||||
var numReg = regexp.MustCompile(`\D?\d*\.?\d*\D?`)
|
||||
|
||||
// series holds input_series defined in the test file
|
||||
type series struct {
|
||||
@@ -94,18 +94,12 @@ func parseInputSeries(input []series, interval *promutil.Duration, startStamp ti
|
||||
// parseInputValue support input like "1", "1+1x1 _ -4 3+20x1", see more examples in test.
|
||||
func parseInputValue(input string, origin bool) ([]sequenceValue, error) {
|
||||
var res []sequenceValue
|
||||
items := strings.Fields(input)
|
||||
if len(items) == 0 {
|
||||
return nil, fmt.Errorf("values cannot be an empty string")
|
||||
}
|
||||
items := strings.Split(input, " ")
|
||||
for _, item := range items {
|
||||
if item == "stale" {
|
||||
res = append(res, sequenceValue{Value: decimal.StaleNaN})
|
||||
continue
|
||||
}
|
||||
if strings.Contains(item, "stale") {
|
||||
return nil, fmt.Errorf("stale metric doesn't support operations")
|
||||
}
|
||||
vals := numReg.FindAllString(item, -1)
|
||||
switch len(vals) {
|
||||
case 1:
|
||||
|
||||
@@ -1,7 +1,6 @@
|
||||
package unittest
|
||||
|
||||
import (
|
||||
"math"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
@@ -13,20 +12,17 @@ func TestParseInputValue_Failure(t *testing.T) {
|
||||
f := func(input string) {
|
||||
t.Helper()
|
||||
|
||||
val, err := parseInputValue(input, true)
|
||||
_, err := parseInputValue(input, true)
|
||||
if err == nil {
|
||||
t.Fatalf("expecting non-nil error, got value %+v", val)
|
||||
t.Fatalf("expecting non-nil error")
|
||||
}
|
||||
}
|
||||
|
||||
f("")
|
||||
f("x4")
|
||||
f("testfailed")
|
||||
|
||||
// stale doesn't support operations
|
||||
f("stalex3")
|
||||
|
||||
f("1+stalex3")
|
||||
}
|
||||
|
||||
func TestParseInputValue_Success(t *testing.T) {
|
||||
@@ -46,8 +42,7 @@ func TestParseInputValue_Success(t *testing.T) {
|
||||
t.Fatalf("unexpected Omitted field in the output\ngot\n%v\nwant\n%v", output, outputExpected)
|
||||
}
|
||||
if outputExpected[i].Value != output[i].Value {
|
||||
if (math.IsNaN(outputExpected[i].Value) && math.IsNaN(output[i].Value)) ||
|
||||
(decimal.IsStaleNaN(outputExpected[i].Value) && decimal.IsStaleNaN(output[i].Value)) {
|
||||
if decimal.IsStaleNaN(outputExpected[i].Value) && decimal.IsStaleNaN(output[i].Value) {
|
||||
continue
|
||||
}
|
||||
t.Fatalf("unexpected Value field in the output\ngot\n%v\nwant\n%v", output, outputExpected)
|
||||
@@ -61,8 +56,6 @@ func TestParseInputValue_Success(t *testing.T) {
|
||||
|
||||
f("stale", []sequenceValue{{Value: decimal.StaleNaN}})
|
||||
|
||||
f(" stale 2", []sequenceValue{{Value: decimal.StaleNaN}, {Value: 2}})
|
||||
|
||||
f("-4x1", []sequenceValue{{Value: -4}, {Value: -4}})
|
||||
|
||||
f("_x1", []sequenceValue{{Omitted: true}})
|
||||
@@ -72,14 +65,6 @@ func TestParseInputValue_Success(t *testing.T) {
|
||||
f("2-1x4", []sequenceValue{{Value: 2}, {Value: 1}, {Value: 0}, {Value: -1}, {Value: -2}})
|
||||
|
||||
f("1+1x1 _ -4 stale 3+20x1", []sequenceValue{{Value: 1}, {Value: 2}, {Omitted: true}, {Value: -4}, {Value: decimal.StaleNaN}, {Value: 3}, {Value: 23}})
|
||||
|
||||
f("Inf +Inf -Inf", []sequenceValue{{Value: math.Inf(1)}, {Value: math.Inf(1)}, {Value: math.Inf(-1)}})
|
||||
|
||||
f("Nan Infx2", []sequenceValue{{Value: math.NaN()}, {Value: math.Inf(1)}, {Value: math.Inf(1)}, {Value: math.Inf(1)}})
|
||||
|
||||
f("NaNx2", []sequenceValue{{Value: math.NaN()}, {Value: math.NaN()}, {Value: math.NaN()}})
|
||||
|
||||
f("4e9+1x2 -3e-2x2 1e-2-1x1", []sequenceValue{{Value: 4000000000}, {Value: 4000000001}, {Value: 4000000002}, {Value: -0.03}, {Value: -0.03}, {Value: -0.03}, {Value: 0.01}, {Value: -0.99}})
|
||||
}
|
||||
|
||||
func TestParseInputSeries_Success(t *testing.T) {
|
||||
|
||||
@@ -13,6 +13,7 @@ import (
|
||||
"path/filepath"
|
||||
"reflect"
|
||||
"sort"
|
||||
"strconv"
|
||||
"strings"
|
||||
"syscall"
|
||||
"time"
|
||||
@@ -98,11 +99,8 @@ func UnitTest(files []string, disableGroupLabel bool, externalLabels []string, e
|
||||
}()
|
||||
}
|
||||
|
||||
tmpFolder, err := os.MkdirTemp(os.TempDir(), testStoragePath)
|
||||
if err != nil {
|
||||
logger.Fatalf("failed to create tmp dir for tests: %v", err)
|
||||
}
|
||||
storagePath = tmpFolder
|
||||
// adding time.Now().UnixNano() to avoid possible file conflict when multiple processes run on a single host
|
||||
storagePath = filepath.Join(os.TempDir(), testStoragePath, strconv.FormatInt(time.Now().UnixNano(), 10))
|
||||
processFlags()
|
||||
vminsert.Init()
|
||||
vmselect.Init()
|
||||
|
||||
@@ -74,7 +74,7 @@ test-vmalert:
|
||||
go test -v -race -cover ./app/vmalert/notifier
|
||||
go test -v -race -cover ./app/vmalert/config
|
||||
go test -v -race -cover ./app/vmalert/remotewrite
|
||||
go test -v -race -cover ./app/vmalert/vmalertutil
|
||||
go test -v -race -cover ./app/vmalert/utils
|
||||
|
||||
run-vmalert: vmalert
|
||||
./bin/vmalert -rule=app/vmalert/config/testdata/rules/rules2-good.rules \
|
||||
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user