vmalert: prepare package for external usage

* update README according to changes
* add Makefile with basic commands
This commit is contained in:
hagen1778 2020-04-12 12:47:26 +01:00 committed by Artem Navoiev
parent 948f8b6b5f
commit 2eed6c393f
6 changed files with 173 additions and 36 deletions

78
app/vmalert/Makefile Normal file
View File

@ -0,0 +1,78 @@
# All these commands must run from repository root.
vmalert:
APP_NAME=vmalert $(MAKE) app-local
vmalert-race:
APP_NAME=vmalert RACE=-race $(MAKE) app-local
vmalert-prod:
APP_NAME=vmalert $(MAKE) app-via-docker
vmalert-pure-prod:
APP_NAME=vmalert $(MAKE) app-via-docker-pure
vmalert-amd64-prod:
APP_NAME=vmalert $(MAKE) app-via-docker-amd64
vmalert-arm-prod:
APP_NAME=vmalert $(MAKE) app-via-docker-arm
vmalert-arm64-prod:
APP_NAME=vmalert $(MAKE) app-via-docker-arm64
vmalert-ppc64le-prod:
APP_NAME=vmalert $(MAKE) app-via-docker-ppc64le
vmalert-386-prod:
APP_NAME=vmalert $(MAKE) app-via-docker-386
package-vmalert:
APP_NAME=vmalert $(MAKE) package-via-docker
package-vmalert-pure:
APP_NAME=vmalert $(MAKE) package-via-docker-pure
package-vmalert-amd64:
APP_NAME=vmalert $(MAKE) package-via-docker-amd64
package-vmalert-arm:
APP_NAME=vmalert $(MAKE) package-via-docker-arm
package-vmalert-arm64:
APP_NAME=vmalert $(MAKE) package-via-docker-arm64
package-vmalert-ppc64le:
APP_NAME=vmalert $(MAKE) package-via-docker-ppc64le
package-vmalert-386:
APP_NAME=vmalert $(MAKE) package-via-docker-386
publish-vmalert:
APP_NAME=vmalert $(MAKE) publish-via-docker
test-vmalert:
go test -race -cover ./app/vmalert
run-vmalert: vmalert
./bin/vmalert -rule=app/vmalert/testdata/rules0-good.rules \
-datasource.url=http://localhost:8428 -notifier.url=http://localhost:9093 \
-evaluationInterval=3s
vmalert-amd64:
CGO_ENABLED=1 GOOS=linux GOARCH=amd64 GO111MODULE=on go build -mod=vendor -ldflags "$(GO_BUILDINFO)" -o bin/vmalert-amd64 ./app/vmalert
vmalert-arm:
CGO_ENABLED=0 GOOS=linux GOARCH=arm GO111MODULE=on go build -mod=vendor -ldflags "$(GO_BUILDINFO)" -o bin/vmalert-arm ./app/vmalert
vmalert-arm64:
CGO_ENABLED=0 GOOS=linux GOARCH=arm64 GO111MODULE=on go build -mod=vendor -ldflags "$(GO_BUILDINFO)" -o bin/vmalert-arm64 ./app/vmalert
vmalert-ppc64le:
CGO_ENABLED=0 GOOS=linux GOARCH=ppc64le GO111MODULE=on go build -mod=vendor -ldflags "$(GO_BUILDINFO)" -o bin/vmalert-ppc64le ./app/vmalert
vmalert-386:
CGO_ENABLED=0 GOOS=linux GOARCH=386 GO111MODULE=on go build -mod=vendor -ldflags "$(GO_BUILDINFO)" -o bin/vmalert-386 ./app/vmalert
vmalert-pure:
APP_NAME=vmalert $(MAKE) app-local-pure

View File

@ -1,41 +1,91 @@
## VM Alert
#### Abstract
The application which accepts the alert rules, executes them on given source, sends(fires) an alert to(in) alert management system
`vmalert` executes a list of given MetricsQL expressions (rules) and
sends alerts to [Alert Manager](https://github.com/prometheus/alertmanager).
### Components
NOTE: `vmalert` is in early alpha and wasn't tested in production systems yet.
#### Alert Config Reader
It accepts yaml config as input parameter in Prometheus format, parses it into Go struct.
### Features:
* Integration with [VictoriaMetrics](https://github.com/VictoriaMetrics/VictoriaMetrics) TSDB;
* VictoriaMetrics [MetricsQL](https://github.com/VictoriaMetrics/VictoriaMetrics/wiki/MetricsQL)
expressions validation;
* Prometheus [alerting rules definition format](https://prometheus.io/docs/prometheus/latest/configuration/alerting_rules/#defining-alerting-rules)
support;
* Integration with [Alertmanager](https://github.com/prometheus/alertmanager);
* Lightweight without extra dependencies.
#### Source Caller
Create own watchdog for every alert group (goroutines), which executes alert query on given source and issues an alert if source returns non-empty result.
Source can be any service which supports PromQL (MetricsQL).
### TODO:
* Persist alerts state as timeseries in TSDB. Currently, alerts state is stored
in process memory only and will be lost on restart;
* Configuration hot reload.
#### Alert Management System Provider
Send positive alert to alert management system, provides interface for every concrete implementation.
Should be ingratiated with Prometheus alertmanager.
### QuickStart
open questions:
- do we really need alert group or can just run every alert in own goroutine?
To build `vmalert` from sources:
```
git clone https://github.com/VictoriaMetrics/VictoriaMetrics
cd VictoriaMetrics
make vmalert
```
The build binary will be placed to `VictoriaMetrics/bin` folder.
#### Web Server
Expose metrics
To start using `vmalert` you will need the following things:
* list of alert rules - PromQL/MetricsQL expressions to execute;
* datasource address - reachable VictoriaMetrics instance for rules execution;
* notifier address - reachable Alertmanager instance for processing,
aggregating alerts and sending notifications.
open questions:
- should the tool provide API or UI for managing alerting rules? Where to store config updated via the API or UI?
- should the tool provide “alerting rules validation mode” for validating and debugging alerting rules? This mode is useful when creating and debugging alerting rules.
Then configure `vmalert` accordingly:
```
./bin/vmalert -rule=alert.rules \
-datasource.url=http://localhost:8428 \
-notifier.url=http://localhost:9093
```
#### Requirements:
- Stateless
- Avoid external dependencies if possible
- Reuse existing code from VictoriaMetrics repo
- Makefile rules for common tasks see Makefiles for other apps in the app/ dir
- Every package should be covered by tests
- Dockerfile
- Graceful shutdown
- Helm template
- Application uses command line flags for configuration
Example for `.rules` file bay be found [here](https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/app/vmalert/testdata/rules0-good.rules)
`vmalert` runs evaluation for every group in a separate goroutine.
Rules in group evaluated one-by-one sequentially.
<img alt="VM Alert" src="vmalert.png">
`vmalert` also runs a web-server (`-httpListenAddr`) for serving metrics and alerts endpoints:
* `http://<vmalert-addr>/api/v1/alerts` - list of all active alerts;
* `http://<vmalert-addr>/api/v1/<groupName>/<alertID>/status" ` - get alert status by ID.
Used as alert source in AlertManager.
* `http://<vmalert-addr>/metrics` - application metrics.
### Configuration
The shortlist of configuration flags is the following:
```
Usage of vmalert:
-datasource.url string
Victoria Metrics or VMSelect url. Required parameter. e.g. http://127.0.0.1:8428
-datasource.basicAuth.password string
Optional basic auth password to use for -datasource.url
-datasource.basicAuth.username string
Optional basic auth username to use for -datasource.url
-evaluationInterval duration
How often to evaluate the rules. Default 1m (default 1m0s)
-external.url string
External URL is used as alert's source for sent alerts to the notifier
-notifier.url string
Prometheus alertmanager URL. Required parameter. e.g. http://127.0.0.1:9093
-rule value
Path to the file with alert rules.
Supports patterns. Flag can be specified multiple times.
Examples:
-rule /path/to/file. Path to a single file with alerting rules
-rule dir/*.yaml -rule /*.yaml. Relative path to all .yaml files in "dir" folder,
absolute path to all .yaml files in root.
-rule.validateAnnotations
Indicates to validate annotation templates (default true)
```
Pass `-help` to `vmalert` in order to see the full list of supported
command-line flags with their descriptions.
### Contributing
`vmalert` is mostly designed and built by VictoriaMetrics community.
Feel free to share your experience and ideas for improving this
software. Please keep simplicity as the main priority.

View File

@ -23,11 +23,12 @@ import (
)
var (
rulePath = flagutil.NewArray("rule", `Path to file with alert rules, accepts patterns.
Flag can be specified multiple time.
rulePath = flagutil.NewArray("rule", `Path to the file with alert rules.
Supports patterns. Flag can be specified multiple times.
Examples:
-rule /path/to/file. Path to single file with alerting rules
-rule dir/*.yaml -rule /*.yaml. Paths to all yaml files in relative dir folder and absolute yaml file in a root.`)
-rule /path/to/file. Path to a single file with alerting rules
-rule dir/*.yaml -rule /*.yaml. Relative path to all .yaml files in "dir" folder,
absolute path to all .yaml files in root.`)
validateAlertAnnotations = flag.Bool("rule.validateAnnotations", true, "Indicates to validate annotation templates")
httpListenAddr = flag.String("httpListenAddr", ":8880", "Address to listen for http connections")
datasourceURL = flag.String("datasource.url", "", "Victoria Metrics or VMSelect url. Required parameter. e.g. http://127.0.0.1:8428")
@ -35,7 +36,7 @@ Examples:
basicAuthPassword = flag.String("datasource.basicAuth.password", "", "Optional basic auth password to use for -datasource.url")
evaluationInterval = flag.Duration("evaluationInterval", 1*time.Minute, "How often to evaluate the rules. Default 1m")
notifierURL = flag.String("notifier.url", "", "Prometheus alertmanager URL. Required parameter. e.g. http://127.0.0.1:9093")
externalURL = flag.String("external.url", "", "URL is used to generate sharable alert URL")
externalURL = flag.String("external.url", "", "External URL is used as alert's source for sent alerts to the notifier")
)
// TODO: hot configuration reload

View File

@ -12,7 +12,6 @@ import (
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/datasource"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmalert/notifier"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/metricsql"
"github.com/VictoriaMetrics/metrics"
)
@ -137,7 +136,6 @@ func (r *Rule) Send(_ context.Context, ap notifier.Notifier) error {
r.mu.Unlock()
if len(alertsCopy) < 1 {
logger.Infof("no alerts to send")
return nil
}
alertsSent.Add(len(alertsCopy))

View File

@ -10,3 +10,13 @@ groups:
summary: "{{ $value|humanize }}"
description: "{{$labels}}"
- name: TestGroup
rules:
- alert: Conns
expr: sum(vm_tcplistener_conns) by(instance) > 1
annotations:
summary: "Too high connection number for {{$labels.instance}}"
description: "It is {{ $value }} connections for {{$labels.instance}}"
- alert: ExampleAlertAlwaysFiring
expr: sum by(job)
(up == 1)

Binary file not shown.

Before

Width:  |  Height:  |  Size: 25 KiB