Compare commits
4 Commits
master
...
remove-py-
Author | SHA1 | Date | |
---|---|---|---|
cea16606ba | |||
a03f430d0e | |||
f3f7c215a7 | |||
c75302bdb8 |
52
.drone.yml
52
.drone.yml
@ -4,7 +4,7 @@ name: test
|
||||
|
||||
steps:
|
||||
- name: test
|
||||
image: golang:1.20
|
||||
image: golang:1.17
|
||||
environment:
|
||||
VERSION: ${DRONE_TAG:-${DRONE_COMMIT}}
|
||||
commands:
|
||||
@ -30,7 +30,7 @@ trigger:
|
||||
|
||||
steps:
|
||||
- name: build all binaries
|
||||
image: golang:1.20
|
||||
image: golang:1.17
|
||||
environment:
|
||||
VERSION: ${DRONE_TAG:-${DRONE_COMMIT}}
|
||||
commands:
|
||||
@ -60,15 +60,51 @@ steps:
|
||||
when:
|
||||
event: tag
|
||||
|
||||
- name: Build and publish docker images
|
||||
image: thegeeklab/drone-docker-buildx
|
||||
- name: push image - arm
|
||||
image: plugins/docker
|
||||
settings:
|
||||
repo: iamthefij/minitor-go
|
||||
auto_tag: true
|
||||
platforms:
|
||||
- linux/amd64
|
||||
- linux/arm64
|
||||
- linux/arm
|
||||
auto_tag_suffix: linux-arm
|
||||
username:
|
||||
from_secret: docker_username
|
||||
password:
|
||||
from_secret: docker_password
|
||||
build_args:
|
||||
- ARCH=arm
|
||||
- REPO=arm32v7
|
||||
|
||||
- name: push image - arm64
|
||||
image: plugins/docker
|
||||
settings:
|
||||
repo: iamthefij/minitor-go
|
||||
auto_tag: true
|
||||
auto_tag_suffix: linux-arm64
|
||||
username:
|
||||
from_secret: docker_username
|
||||
password:
|
||||
from_secret: docker_password
|
||||
build_args:
|
||||
- ARCH=arm64
|
||||
- REPO=arm64v8
|
||||
|
||||
- name: push image - amd64
|
||||
image: plugins/docker
|
||||
settings:
|
||||
repo: iamthefij/minitor-go
|
||||
auto_tag: true
|
||||
auto_tag_suffix: linux-amd64
|
||||
username:
|
||||
from_secret: docker_username
|
||||
password:
|
||||
from_secret: docker_password
|
||||
|
||||
- name: publish manifest
|
||||
image: plugins/manifest
|
||||
settings:
|
||||
spec: manifest.tmpl
|
||||
auto_tag: true
|
||||
ignore_missing: true
|
||||
username:
|
||||
from_secret: docker_username
|
||||
password:
|
||||
|
10
Dockerfile
10
Dockerfile
@ -1,10 +1,11 @@
|
||||
FROM alpine:3.18
|
||||
ARG REPO=library
|
||||
FROM ${REPO}/alpine:3.17
|
||||
|
||||
RUN mkdir /app
|
||||
WORKDIR /app/
|
||||
|
||||
# Add common checking tools
|
||||
RUN apk --no-cache add bash=~5 curl=~8 jq=~1 bind-tools=~9 tzdata~=2024a
|
||||
RUN apk --no-cache add bash=~5 curl=~8 jq=~1.6 bind-tools~=9
|
||||
|
||||
# Add minitor user for running as non-root
|
||||
RUN addgroup -S minitor && adduser -S minitor -G minitor
|
||||
@ -14,9 +15,8 @@ COPY ./scripts /app/scripts
|
||||
RUN chmod -R 755 /app/scripts
|
||||
|
||||
# Copy minitor in
|
||||
ARG TARGETOS
|
||||
ARG TARGETARCH
|
||||
COPY ./dist/minitor-${TARGETOS}-${TARGETARCH} ./minitor
|
||||
ARG ARCH=amd64
|
||||
COPY ./dist/minitor-linux-${ARCH} ./minitor
|
||||
|
||||
# Drop to non-root user
|
||||
USER minitor
|
||||
|
@ -1,4 +1,5 @@
|
||||
FROM golang:1.20 AS builder
|
||||
ARG REPO=library
|
||||
FROM golang:1.17 AS builder
|
||||
|
||||
RUN mkdir /app
|
||||
WORKDIR /app
|
||||
@ -8,13 +9,12 @@ RUN go mod download
|
||||
|
||||
COPY ./*.go /app/
|
||||
|
||||
ARG TARGETOS
|
||||
ARG TARGETARCH
|
||||
ARG ARCH=amd64
|
||||
ARG VERSION=dev
|
||||
ENV CGO_ENABLED=0 GOOS=$TARGETOS GOARCH=${TARGETARCH}
|
||||
ENV CGO_ENABLED=0 GOOS=linux GOARCH=${ARCH}
|
||||
RUN go build -ldflags "-X main.version=${VERSION}" -a -installsuffix nocgo -o minitor .
|
||||
|
||||
FROM alpine:3.18
|
||||
FROM ${REPO}/alpine:3.17
|
||||
RUN mkdir /app
|
||||
WORKDIR /app/
|
||||
|
||||
@ -22,7 +22,7 @@ WORKDIR /app/
|
||||
COPY --from=builder /app/minitor .
|
||||
|
||||
# Add common checking tools
|
||||
RUN apk --no-cache add bash=~5 curl=~8 jq=~1 bind-tools=~9 tzdata~=2024a
|
||||
RUN apk --no-cache add bash=~5 curl=~8 jq=~1.6 bind-tools~=9
|
||||
|
||||
# Add minitor user for running as non-root
|
||||
RUN addgroup -S minitor && adduser -S minitor -G minitor
|
||||
|
4
Makefile
4
Makefile
@ -79,11 +79,11 @@ $(TARGET_ALIAS):
|
||||
# Arch specific docker build targets
|
||||
.PHONY: docker-build-arm
|
||||
docker-build-arm: dist/minitor-linux-arm
|
||||
docker build --platform linux/arm . -t ${DOCKER_TAG}-linux-arm
|
||||
docker build --build-arg REPO=arm32v7 --build-arg ARCH=arm . -t ${DOCKER_TAG}-linux-arm
|
||||
|
||||
.PHONY: docker-build-arm64
|
||||
docker-build-arm64: dist/minitor-linux-arm64
|
||||
docker build --platform linux/arm64 . -t ${DOCKER_TAG}-linux-arm64
|
||||
docker build --build-arg REPO=arm64v8 --build-arg ARCH=arm64 . -t ${DOCKER_TAG}-linux-arm64
|
||||
|
||||
# Cross run on host architechture
|
||||
.PHONY: docker-run-arm
|
||||
|
34
README.md
34
README.md
@ -46,8 +46,6 @@ docker run -v $PWD/config.yml:/app/config.yml iamthefij/minitor-go:latest
|
||||
|
||||
Images are provided for `amd64`, `arm`, and `arm64` architechtures.
|
||||
|
||||
Timezone configuration for the container is set by passing the `TZ` env variable. Eg. `TZ=America/Los_Angeles`.
|
||||
|
||||
## Configuring
|
||||
|
||||
In this repo, you can explore the `sample-config.yml` file for an example, but the general structure is as follows. It should be noted that environment variable interpolation happens on load of the YAML file.
|
||||
@ -96,38 +94,10 @@ Also, when alerts are executed, they will be passed through Go's format function
|
||||
|`{{.AlertCount}}`|Number of times this monitor has alerted|
|
||||
|`{{.FailureCount}}`|The total number of sequential failed checks for this monitor|
|
||||
|`{{.LastCheckOutput}}`|The last returned value from the check command to either stderr or stdout|
|
||||
|`{{.LastSuccess}}`|The datetime of the last successful check as a go Time struct|
|
||||
|`{{.LastSuccess}}`|The ISO datetime of the last successful check|
|
||||
|`{{.MonitorName}}`|The name of the monitor that failed and triggered the alert|
|
||||
|`{{.IsUp}}`|Indicates if the monitor that is alerting is up or not. Can be used in a conditional message template|
|
||||
|
||||
To provide flexible formatting, the following non-standard functions are available in templates:
|
||||
|
||||
|func|description|
|
||||
|---|---|
|
||||
|`ANSIC <Time>`|Formats provided time in ANSIC format|
|
||||
|`UnixDate <Time>`|Formats provided time in UnixDate format|
|
||||
|`RubyDate <Time>`|Formats provided time in RubyDate format|
|
||||
|`RFC822Z <Time>`|Formats provided time in RFC822Z format|
|
||||
|`RFC850 <Time>`|Formats provided time in RFC850 format|
|
||||
|`RFC1123 <Time>`|Formats provided time in RFC1123 format|
|
||||
|`RFC1123Z <Time>`|Formats provided time in RFC1123Z format|
|
||||
|`RFC3339 <Time>`|Formats provided time in RFC3339 format|
|
||||
|`RFC3339Nano <Time>`|Formats provided time in RFC3339Nano format|
|
||||
|`FormatTime <Time> <string template>`|Formats provided time according to provided template|
|
||||
|`InTZ <Time> <string timezone name>`|Converts provided time to parsed timezone from the provided name|
|
||||
|
||||
For more information, check out the [Go documentation for the time module](https://pkg.go.dev/time@go1.20.7#pkg-constants).
|
||||
|
||||
#### Running alerts on startup
|
||||
|
||||
It's not the best feeling to find out your alerts are broken when you're expecting to be alerted about another failure. To avoid this and provide early insight into broken alerts, it is possible to specify a list of alerts to run when Minitor starts up. This can be done using the command line flag `-startup-alerts`. This flag accepts a comma separated list of strings and will run a test of each of those alerts. Minitor will then respond as it typically does for any failed alert. This can be used to allow you time to correct when initially launching, and to allow schedulers to more easily detect a failed deployment of Minitor.
|
||||
|
||||
Eg.
|
||||
|
||||
```bash
|
||||
minitor -startup-alerts=log_down,log_up -config ./config.yml
|
||||
```
|
||||
|
||||
### Metrics
|
||||
|
||||
Minitor supports exporting metrics for [Prometheus](https://prometheus.io/). Prometheus is an open source tool for reading and querying metrics from different sources. Combined with another tool, [Grafana](https://grafana.com/), it allows building of charts and dashboards. You could also opt to just use Minitor to log check results, and instead do your alerting with Grafana.
|
||||
@ -188,7 +158,7 @@ minitor-go:
|
||||
check_interval: 1m30s
|
||||
```
|
||||
|
||||
For the time being, legacy configs for the Python version of Minitor should be compatible if you apply the `-py-compat` flag when running Minitor. Eventually, this flag will go away when later breaking changes are introduced.
|
||||
The `-py-compat` flag has been removed. Any existing Python oriented configuration needs to be migrated to the new templates.
|
||||
|
||||
## Future
|
||||
|
||||
|
56
alert.go
56
alert.go
@ -5,7 +5,6 @@ import (
|
||||
"errors"
|
||||
"fmt"
|
||||
"os/exec"
|
||||
"strings"
|
||||
"text/template"
|
||||
"time"
|
||||
|
||||
@ -45,70 +44,21 @@ func (alert Alert) IsValid() bool {
|
||||
|
||||
// BuildTemplates compiles command templates for the Alert
|
||||
func (alert *Alert) BuildTemplates() error {
|
||||
// TODO: Remove legacy template support later after 1.0
|
||||
legacy := strings.NewReplacer(
|
||||
"{alert_count}", "{{.AlertCount}}",
|
||||
"{alert_message}", "{{.MonitorName}} check has failed {{.FailureCount}} times",
|
||||
"{failure_count}", "{{.FailureCount}}",
|
||||
"{last_output}", "{{.LastCheckOutput}}",
|
||||
"{last_success}", "{{.LastSuccess}}",
|
||||
"{monitor_name}", "{{.MonitorName}}",
|
||||
)
|
||||
|
||||
slog.Debugf("Building template for alert %s", alert.Name)
|
||||
|
||||
// Time format func factory
|
||||
tff := func(formatString string) func(time.Time) string {
|
||||
return func(t time.Time) string {
|
||||
return t.Format(formatString)
|
||||
}
|
||||
}
|
||||
|
||||
// Create some functions for formatting datetimes in popular formats
|
||||
timeFormatFuncs := template.FuncMap{
|
||||
"ANSIC": tff(time.ANSIC),
|
||||
"UnixDate": tff(time.UnixDate),
|
||||
"RubyDate": tff(time.RubyDate),
|
||||
"RFC822Z": tff(time.RFC822Z),
|
||||
"RFC850": tff(time.RFC850),
|
||||
"RFC1123": tff(time.RFC1123),
|
||||
"RFC1123Z": tff(time.RFC1123Z),
|
||||
"RFC3339": tff(time.RFC3339),
|
||||
"RFC3339Nano": tff(time.RFC3339Nano),
|
||||
"FormatTime": func(t time.Time, timeFormat string) string {
|
||||
return t.Format(timeFormat)
|
||||
},
|
||||
"InTZ": func(t time.Time, tzName string) (time.Time, error) {
|
||||
tz, err := time.LoadLocation(tzName)
|
||||
if err != nil {
|
||||
return t, fmt.Errorf("failed to convert time to specified tz: %w", err)
|
||||
}
|
||||
|
||||
return t.In(tz), nil
|
||||
},
|
||||
}
|
||||
|
||||
switch {
|
||||
case alert.commandTemplate == nil && alert.Command.Command != nil:
|
||||
alert.commandTemplate = []*template.Template{}
|
||||
for i, cmdPart := range alert.Command.Command {
|
||||
if PyCompat {
|
||||
cmdPart = legacy.Replace(cmdPart)
|
||||
}
|
||||
|
||||
alert.commandTemplate = append(alert.commandTemplate, template.Must(
|
||||
template.New(alert.Name+fmt.Sprint(i)).Funcs(timeFormatFuncs).Parse(cmdPart),
|
||||
template.New(alert.Name+fmt.Sprint(i)).Parse(cmdPart),
|
||||
))
|
||||
}
|
||||
case alert.commandShellTemplate == nil && alert.Command.ShellCommand != "":
|
||||
shellCmd := alert.Command.ShellCommand
|
||||
|
||||
if PyCompat {
|
||||
shellCmd = legacy.Replace(shellCmd)
|
||||
}
|
||||
|
||||
alert.commandShellTemplate = template.Must(
|
||||
template.New(alert.Name).Funcs(timeFormatFuncs).Parse(shellCmd),
|
||||
template.New(alert.Name).Parse(shellCmd),
|
||||
)
|
||||
default:
|
||||
return fmt.Errorf("No template provided for alert %s: %w", alert.Name, errNoTemplate)
|
||||
@ -168,7 +118,7 @@ func (alert Alert) Send(notice AlertNotice) (outputStr string, err error) {
|
||||
|
||||
if err != nil {
|
||||
err = fmt.Errorf(
|
||||
"Alert %s failed to send. Returned %w: %w",
|
||||
"Alert '%s' failed to send. Returned %v: %w",
|
||||
alert.Name,
|
||||
err,
|
||||
ErrAlertFailed,
|
||||
|
@ -70,14 +70,6 @@ func TestAlertSend(t *testing.T) {
|
||||
"Command shell with bad template",
|
||||
false,
|
||||
},
|
||||
{
|
||||
Alert{Command: CommandOrShell{ShellCommand: "echo {alert_message}"}},
|
||||
AlertNotice{MonitorName: "test", FailureCount: 1},
|
||||
"test check has failed 1 times\n",
|
||||
false,
|
||||
"Command shell with legacy template",
|
||||
true,
|
||||
},
|
||||
// Test default log alert down
|
||||
{
|
||||
*NewLogAlert(),
|
||||
@ -100,8 +92,6 @@ func TestAlertSend(t *testing.T) {
|
||||
|
||||
for _, c := range cases {
|
||||
log.Printf("Testing case %s", c.name)
|
||||
// Set PyCompat to value of compat flag
|
||||
PyCompat = c.pyCompat
|
||||
|
||||
err := c.alert.BuildTemplates()
|
||||
if err != nil {
|
||||
@ -121,9 +111,6 @@ func TestAlertSend(t *testing.T) {
|
||||
log.Printf("Case failed: %s", c.name)
|
||||
}
|
||||
|
||||
// Set PyCompat back to default value
|
||||
PyCompat = false
|
||||
|
||||
log.Println("-----")
|
||||
}
|
||||
}
|
||||
|
50
config.go
50
config.go
@ -13,11 +13,11 @@ var errInvalidConfig = errors.New("Invalid configuration")
|
||||
|
||||
// Config type is contains all provided user configuration
|
||||
type Config struct {
|
||||
CheckInterval SecondsOrDuration `yaml:"check_interval"`
|
||||
DefaultAlertAfter int16 `yaml:"default_alert_after"`
|
||||
DefaultAlertEvery *int16 `yaml:"default_alert_every"`
|
||||
DefaultAlertDown []string `yaml:"default_alert_down"`
|
||||
DefaultAlertUp []string `yaml:"default_alert_up"`
|
||||
CheckInterval time.Duration `yaml:"check_interval"`
|
||||
DefaultAlertAfter int16 `yaml:"default_alert_after"`
|
||||
DefaultAlertEvery *int16 `yaml:"default_alert_every"`
|
||||
DefaultAlertDown []string `yaml:"default_alert_down"`
|
||||
DefaultAlertUp []string `yaml:"default_alert_up"`
|
||||
Monitors []*Monitor
|
||||
Alerts map[string]*Alert
|
||||
}
|
||||
@ -56,34 +56,6 @@ func (cos *CommandOrShell) UnmarshalYAML(unmarshal func(interface{}) error) erro
|
||||
return nil
|
||||
}
|
||||
|
||||
// SecondsOrDuration wraps a duration value for parsing a duration or seconds from YAML
|
||||
// NOTE: This should be removed in favor of only parsing durations once compatibility is broken
|
||||
type SecondsOrDuration struct {
|
||||
value time.Duration
|
||||
}
|
||||
|
||||
// Value returns a duration value
|
||||
func (sod SecondsOrDuration) Value() time.Duration {
|
||||
return sod.value
|
||||
}
|
||||
|
||||
// UnmarshalYAML allows unmarshalling a duration value or seconds if an int was provided
|
||||
func (sod *SecondsOrDuration) UnmarshalYAML(unmarshal func(interface{}) error) error {
|
||||
var seconds int64
|
||||
err := unmarshal(&seconds)
|
||||
|
||||
if err == nil {
|
||||
sod.value = time.Second * time.Duration(seconds)
|
||||
|
||||
return nil
|
||||
}
|
||||
|
||||
// Error indicates that we don't have an int
|
||||
err = unmarshal(&sod.value)
|
||||
|
||||
return err
|
||||
}
|
||||
|
||||
// IsValid checks config validity and returns true if valid
|
||||
func (config Config) IsValid() (isValid bool) {
|
||||
isValid = true
|
||||
@ -182,18 +154,6 @@ func LoadConfig(filePath string) (config Config, err error) {
|
||||
|
||||
slog.Debugf("Config values:\n%v\n", config)
|
||||
|
||||
// Add log alert if not present
|
||||
if PyCompat {
|
||||
// Initialize alerts list if not present
|
||||
if config.Alerts == nil {
|
||||
config.Alerts = map[string]*Alert{}
|
||||
}
|
||||
|
||||
if _, ok := config.Alerts["log"]; !ok {
|
||||
config.Alerts["log"] = NewLogAlert()
|
||||
}
|
||||
}
|
||||
|
||||
// Finish initializing configuration
|
||||
if err = config.Init(); err != nil {
|
||||
return
|
||||
|
@ -15,7 +15,6 @@ func TestLoadConfig(t *testing.T) {
|
||||
}{
|
||||
{"./test/valid-config.yml", false, "Valid config file", false},
|
||||
{"./test/valid-config-default-values.yml", false, "Valid config file with default values", false},
|
||||
{"./test/valid-default-log-alert.yml", false, "Valid config file with default log alert PyCompat", true},
|
||||
{"./test/valid-default-log-alert.yml", true, "Invalid config file no log alert", false},
|
||||
{"./test/does-not-exist", true, "Invalid config path", false},
|
||||
{"./test/invalid-config-type.yml", true, "Invalid config type for key", false},
|
||||
@ -25,8 +24,6 @@ func TestLoadConfig(t *testing.T) {
|
||||
|
||||
for _, c := range cases {
|
||||
log.Printf("Testing case %s", c.name)
|
||||
// Set PyCompat based on compatibility mode
|
||||
PyCompat = c.pyCompat
|
||||
_, err := LoadConfig(c.configPath)
|
||||
hasErr := (err != nil)
|
||||
|
||||
@ -34,9 +31,6 @@ func TestLoadConfig(t *testing.T) {
|
||||
t.Errorf("LoadConfig(%v), expected_error=%v actual=%v", c.name, c.expectErr, err)
|
||||
log.Printf("Case failed: %s", c.name)
|
||||
}
|
||||
|
||||
// Set PyCompat to default value
|
||||
PyCompat = false
|
||||
}
|
||||
}
|
||||
|
||||
@ -53,15 +47,15 @@ func TestIntervalParsing(t *testing.T) {
|
||||
oneMinute := time.Minute
|
||||
|
||||
// validate top level interval seconds represented as an int
|
||||
if config.CheckInterval.Value() != oneSecond {
|
||||
if config.CheckInterval != oneSecond {
|
||||
t.Errorf("Incorrectly parsed int seconds. expected=%v actual=%v", oneSecond, config.CheckInterval)
|
||||
}
|
||||
|
||||
if config.Monitors[0].CheckInterval.Value() != tenSeconds {
|
||||
if config.Monitors[0].CheckInterval != tenSeconds {
|
||||
t.Errorf("Incorrectly parsed seconds duration. expected=%v actual=%v", oneSecond, config.CheckInterval)
|
||||
}
|
||||
|
||||
if config.Monitors[1].CheckInterval.Value() != oneMinute {
|
||||
if config.Monitors[1].CheckInterval != oneMinute {
|
||||
t.Errorf("Incorrectly parsed seconds duration. expected=%v actual=%v", oneSecond, config.CheckInterval)
|
||||
}
|
||||
|
||||
|
14
go.mod
14
go.mod
@ -1,10 +1,10 @@
|
||||
module git.iamthefij.com/iamthefij/minitor-go
|
||||
|
||||
go 1.20
|
||||
go 1.17
|
||||
|
||||
require (
|
||||
git.iamthefij.com/iamthefij/slog v1.3.0
|
||||
github.com/prometheus/client_golang v1.19.0
|
||||
github.com/prometheus/client_golang v1.15.0
|
||||
gopkg.in/yaml.v2 v2.4.0
|
||||
)
|
||||
|
||||
@ -14,10 +14,10 @@ require (
|
||||
github.com/golang/protobuf v1.5.3 // indirect
|
||||
github.com/kr/text v0.2.0 // indirect
|
||||
github.com/matttproud/golang_protobuf_extensions v1.0.4 // indirect
|
||||
github.com/prometheus/client_model v0.5.0 // indirect
|
||||
github.com/prometheus/common v0.48.0 // indirect
|
||||
github.com/prometheus/procfs v0.12.0 // indirect
|
||||
github.com/prometheus/client_model v0.3.0 // indirect
|
||||
github.com/prometheus/common v0.42.0 // indirect
|
||||
github.com/prometheus/procfs v0.9.0 // indirect
|
||||
github.com/rogpeppe/go-internal v1.10.0 // indirect
|
||||
golang.org/x/sys v0.16.0 // indirect
|
||||
google.golang.org/protobuf v1.33.0 // indirect
|
||||
golang.org/x/sys v0.6.0 // indirect
|
||||
google.golang.org/protobuf v1.30.0 // indirect
|
||||
)
|
||||
|
14
go.sum
14
go.sum
@ -182,16 +182,12 @@ github.com/prometheus/client_golang v1.12.1/go.mod h1:3Z9XVyYiZYEO+YQWt3RD2R3jrb
|
||||
github.com/prometheus/client_golang v1.14.0/go.mod h1:8vpkKitgIVNcqrRBWh1C4TIUQgYNtG/XQE4E/Zae36Y=
|
||||
github.com/prometheus/client_golang v1.15.0 h1:5fCgGYogn0hFdhyhLbw7hEsWxufKtY9klyvdNfFlFhM=
|
||||
github.com/prometheus/client_golang v1.15.0/go.mod h1:e9yaBhRPU2pPNsZwE+JdQl0KEt1N9XgF6zxWmaC0xOk=
|
||||
github.com/prometheus/client_golang v1.19.0 h1:ygXvpU1AoN1MhdzckN+PyD9QJOSD4x7kmXYlnfbA6JU=
|
||||
github.com/prometheus/client_golang v1.19.0/go.mod h1:ZRM9uEAypZakd+q/x7+gmsvXdURP+DABIEIjnmDdp+k=
|
||||
github.com/prometheus/client_model v0.0.0-20180712105110-5c3871d89910/go.mod h1:MbSGuTsp3dbXC40dX6PRTWyKYBIrTGTE9sqQNg2J8bo=
|
||||
github.com/prometheus/client_model v0.0.0-20190129233127-fd36f4220a90/go.mod h1:xMI15A0UPsDsEKsMN9yxemIoYk6Tm2C1GtYGdfGttqA=
|
||||
github.com/prometheus/client_model v0.0.0-20190812154241-14fe0d1b01d4/go.mod h1:xMI15A0UPsDsEKsMN9yxemIoYk6Tm2C1GtYGdfGttqA=
|
||||
github.com/prometheus/client_model v0.2.0/go.mod h1:xMI15A0UPsDsEKsMN9yxemIoYk6Tm2C1GtYGdfGttqA=
|
||||
github.com/prometheus/client_model v0.3.0 h1:UBgGFHqYdG/TPFD1B1ogZywDqEkwp3fBMvqdiQ7Xew4=
|
||||
github.com/prometheus/client_model v0.3.0/go.mod h1:LDGWKZIo7rky3hgvBe+caln+Dr3dPggB5dvjtD7w9+w=
|
||||
github.com/prometheus/client_model v0.5.0 h1:VQw1hfvPvk3Uv6Qf29VrPF32JB6rtbgI6cYPYQjL0Qw=
|
||||
github.com/prometheus/client_model v0.5.0/go.mod h1:dTiFglRmd66nLR9Pv9f0mZi7B7fk5Pm3gvsjB5tr+kI=
|
||||
github.com/prometheus/common v0.4.1/go.mod h1:TNfzLD0ON7rHzMJeJkieUDPYmFC7Snx/y86RQel1bk4=
|
||||
github.com/prometheus/common v0.10.0/go.mod h1:Tlit/dnDKsSWFlCLTWaA1cyBgKHSMdTB80sz/V91rCo=
|
||||
github.com/prometheus/common v0.26.0/go.mod h1:M7rCNAaPfAosfx8veZJCuw84e35h3Cfd9VFqTh1DIvc=
|
||||
@ -199,8 +195,6 @@ github.com/prometheus/common v0.32.1/go.mod h1:vu+V0TpY+O6vW9J44gczi3Ap/oXXR10b+
|
||||
github.com/prometheus/common v0.37.0/go.mod h1:phzohg0JFMnBEFGxTDbfu3QyL5GI8gTQJFhYO5B3mfA=
|
||||
github.com/prometheus/common v0.42.0 h1:EKsfXEYo4JpWMHH5cg+KOUWeuJSov1Id8zGR8eeI1YM=
|
||||
github.com/prometheus/common v0.42.0/go.mod h1:xBwqVerjNdUDjgODMpudtOMwlOwf2SaTr1yjz4b7Zbc=
|
||||
github.com/prometheus/common v0.48.0 h1:QO8U2CdOzSn1BBsmXJXduaaW+dY/5QLjfB8svtSzKKE=
|
||||
github.com/prometheus/common v0.48.0/go.mod h1:0/KsvlIEfPQCQ5I2iNSAWKPZziNCvRs5EC6ILDTlAPc=
|
||||
github.com/prometheus/procfs v0.0.0-20181005140218-185b4288413d/go.mod h1:c3At6R/oaqEKCNdg8wHV1ftS6bRYblBhIjjI8uT2IGk=
|
||||
github.com/prometheus/procfs v0.0.2/go.mod h1:TjEm7ze935MbeOT/UhFTIMYKhuLP4wbCsTZCD3I8kEA=
|
||||
github.com/prometheus/procfs v0.1.3/go.mod h1:lV6e/gmhEcM9IjHGsFOCxxuZ+z1YqCvr4OA4YeYWdaU=
|
||||
@ -209,8 +203,6 @@ github.com/prometheus/procfs v0.7.3/go.mod h1:cz+aTbrPOrUb4q7XlbU9ygM+/jj0fzG6c1
|
||||
github.com/prometheus/procfs v0.8.0/go.mod h1:z7EfXMXOkbkqb9IINtpCn86r/to3BnA0uaxHdg830/4=
|
||||
github.com/prometheus/procfs v0.9.0 h1:wzCHvIvM5SxWqYvwgVL7yJY8Lz3PKn49KQtpgMYJfhI=
|
||||
github.com/prometheus/procfs v0.9.0/go.mod h1:+pB4zwohETzFnmlpe6yd2lSc+0/46IYZRB/chUwxUZY=
|
||||
github.com/prometheus/procfs v0.12.0 h1:jluTpSng7V9hY0O2R9DzzJHYb2xULk9VTR1V1R/k6Bo=
|
||||
github.com/prometheus/procfs v0.12.0/go.mod h1:pcuDEFsWDnvcgNzo4EEweacyhjeA9Zk3cnaOZAZEfOo=
|
||||
github.com/rogpeppe/go-internal v1.3.0/go.mod h1:M8bDsm7K2OlrFYOpmOWEs/qY81heoFRclV5y23lUDJ4=
|
||||
github.com/rogpeppe/go-internal v1.9.0/go.mod h1:WtVeX8xhTBvf0smdhujwtBcq4Qrzq/fJaraNFVN+nFs=
|
||||
github.com/rogpeppe/go-internal v1.10.0 h1:TMyTOH3F/DB16zRVcYyreMH6GnZZrwQVAoYjRBZyWFQ=
|
||||
@ -373,8 +365,6 @@ golang.org/x/sys v0.3.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
|
||||
golang.org/x/sys v0.5.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
|
||||
golang.org/x/sys v0.6.0 h1:MVltZSvRTcU2ljQOhs94SXPftV6DCNnZViHeQps87pQ=
|
||||
golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
|
||||
golang.org/x/sys v0.16.0 h1:xWw16ngr6ZMtmxDyKyIgsE93KNKz5HKmMa3b8ALHidU=
|
||||
golang.org/x/sys v0.16.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
|
||||
golang.org/x/term v0.0.0-20201126162022-7de9c90e9dd1/go.mod h1:bj7SfCRtBDWHUb9snDiAeCFNEtKQo2Wmx5Cou7ajbmo=
|
||||
golang.org/x/term v0.0.0-20210927222741-03fcf44c2211/go.mod h1:jbD1KX2456YbFQfuXm/mYQcufACuNUgVhRMnK/tPxf8=
|
||||
golang.org/x/term v0.5.0/go.mod h1:jMB1sMXY+tzblOD4FWmEbocvup2/aLOaQEp7JmGp78k=
|
||||
@ -514,10 +504,6 @@ google.golang.org/protobuf v1.28.0/go.mod h1:HV8QOd/L58Z+nl8r43ehVNZIU/HEI6OcFqw
|
||||
google.golang.org/protobuf v1.28.1/go.mod h1:HV8QOd/L58Z+nl8r43ehVNZIU/HEI6OcFqwMG9pJV4I=
|
||||
google.golang.org/protobuf v1.30.0 h1:kPPoIgf3TsEvrm0PFe15JQ+570QVxYzEvvHqChK+cng=
|
||||
google.golang.org/protobuf v1.30.0/go.mod h1:HV8QOd/L58Z+nl8r43ehVNZIU/HEI6OcFqwMG9pJV4I=
|
||||
google.golang.org/protobuf v1.32.0 h1:pPC6BG5ex8PDFnkbrGU3EixyhKcQ2aDuBS36lqK/C7I=
|
||||
google.golang.org/protobuf v1.32.0/go.mod h1:c6P6GXX6sHbq/GpV6MGZEdwhWPcYBgnhAHhKbcUYpos=
|
||||
google.golang.org/protobuf v1.33.0 h1:uNO2rsAINq/JlFpSdYEKIZ0uKD/R9cpdv0T+yoGwGmI=
|
||||
google.golang.org/protobuf v1.33.0/go.mod h1:c6P6GXX6sHbq/GpV6MGZEdwhWPcYBgnhAHhKbcUYpos=
|
||||
gopkg.in/alecthomas/kingpin.v2 v2.2.6/go.mod h1:FMv+mEhP44yOT+4EoQTLFTRgOQ1FBLkstjWtayDeSgw=
|
||||
gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
|
||||
gopkg.in/check.v1 v1.0.0-20180628173108-788fd7840127/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
|
||||
|
48
main.go
48
main.go
@ -4,7 +4,6 @@ import (
|
||||
"errors"
|
||||
"flag"
|
||||
"fmt"
|
||||
"strings"
|
||||
"time"
|
||||
|
||||
"git.iamthefij.com/iamthefij/slog"
|
||||
@ -18,9 +17,6 @@ var (
|
||||
// Metrics contains all active metrics
|
||||
Metrics = NewMetrics()
|
||||
|
||||
// PyCompat enables support for legacy Python templates
|
||||
PyCompat = false
|
||||
|
||||
// version of minitor being run
|
||||
version = "dev"
|
||||
|
||||
@ -92,42 +88,12 @@ func checkMonitors(config *Config) error {
|
||||
return nil
|
||||
}
|
||||
|
||||
func sendStartupAlerts(config *Config, alertNames []string) error {
|
||||
for _, alertName := range alertNames {
|
||||
var err error
|
||||
|
||||
alert, ok := config.Alerts[alertName]
|
||||
if !ok {
|
||||
err = fmt.Errorf("unknown alert %s: %w", alertName, errUnknownAlert)
|
||||
}
|
||||
|
||||
if err == nil {
|
||||
_, err = alert.Send(AlertNotice{
|
||||
AlertCount: 0,
|
||||
FailureCount: 0,
|
||||
IsUp: true,
|
||||
LastSuccess: time.Now(),
|
||||
MonitorName: fmt.Sprintf("First Run Alert Test: %s", alert.Name),
|
||||
LastCheckOutput: "",
|
||||
})
|
||||
}
|
||||
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
}
|
||||
|
||||
return nil
|
||||
}
|
||||
|
||||
func main() {
|
||||
showVersion := flag.Bool("version", false, "Display the version of minitor and exit")
|
||||
configPath := flag.String("config", "config.yml", "Alternate configuration path (default: config.yml)")
|
||||
startupAlerts := flag.String("startup-alerts", "", "List of alerts to run on startup. This can help determine unhealthy alerts early on. (default \"\")")
|
||||
|
||||
flag.BoolVar(&slog.DebugLevel, "debug", false, "Enables debug logs (default: false)")
|
||||
flag.BoolVar(&ExportMetrics, "metrics", false, "Enables prometheus metrics exporting (default: false)")
|
||||
flag.BoolVar(&PyCompat, "py-compat", false, "Enables support for legacy Python Minitor config. Will eventually be removed. (default: false)")
|
||||
flag.IntVar(&MetricsPort, "metrics-port", MetricsPort, "The port that Prometheus metrics should be exported on, if enabled. (default: 8080)")
|
||||
flag.Parse()
|
||||
|
||||
@ -138,10 +104,6 @@ func main() {
|
||||
return
|
||||
}
|
||||
|
||||
if PyCompat {
|
||||
slog.Warningf("Python compatibility mode is enabled. This will be removed in the next major release. Please update your configuration.")
|
||||
}
|
||||
|
||||
// Load configuration
|
||||
config, err := LoadConfig(*configPath)
|
||||
slog.OnErrFatalf(err, "Error loading config: %v", err)
|
||||
@ -153,19 +115,11 @@ func main() {
|
||||
go ServeMetrics()
|
||||
}
|
||||
|
||||
if *startupAlerts != "" {
|
||||
alertNames := strings.Split(*startupAlerts, ",")
|
||||
|
||||
err = sendStartupAlerts(&config, alertNames)
|
||||
|
||||
slog.OnErrPanicf(err, "Error running startup alerts")
|
||||
}
|
||||
|
||||
// Start main loop
|
||||
for {
|
||||
err = checkMonitors(&config)
|
||||
slog.OnErrPanicf(err, "Error checking monitors")
|
||||
|
||||
time.Sleep(config.CheckInterval.Value())
|
||||
time.Sleep(config.CheckInterval)
|
||||
}
|
||||
}
|
||||
|
61
main_test.go
61
main_test.go
@ -134,64 +134,3 @@ func TestCheckMonitors(t *testing.T) {
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func TestFirstRunAlerts(t *testing.T) {
|
||||
cases := []struct {
|
||||
config Config
|
||||
expectErr bool
|
||||
startupAlerts []string
|
||||
name string
|
||||
}{
|
||||
{
|
||||
config: Config{},
|
||||
expectErr: false,
|
||||
startupAlerts: []string{},
|
||||
name: "Empty",
|
||||
},
|
||||
{
|
||||
config: Config{},
|
||||
expectErr: true,
|
||||
startupAlerts: []string{"missing"},
|
||||
name: "Unknown",
|
||||
},
|
||||
{
|
||||
config: Config{
|
||||
Alerts: map[string]*Alert{
|
||||
"good": {
|
||||
Command: CommandOrShell{Command: []string{"true"}},
|
||||
},
|
||||
},
|
||||
},
|
||||
expectErr: false,
|
||||
startupAlerts: []string{"good"},
|
||||
name: "Successful alert",
|
||||
},
|
||||
{
|
||||
config: Config{
|
||||
Alerts: map[string]*Alert{
|
||||
"bad": {
|
||||
Name: "bad",
|
||||
Command: CommandOrShell{Command: []string{"false"}},
|
||||
},
|
||||
},
|
||||
},
|
||||
expectErr: true,
|
||||
startupAlerts: []string{"bad"},
|
||||
name: "Failed alert",
|
||||
},
|
||||
}
|
||||
|
||||
for _, c := range cases {
|
||||
err := c.config.Init()
|
||||
if err != nil {
|
||||
t.Errorf("sendFirstRunAlerts(%s): unexpected error reading config: %v", c.name, err)
|
||||
}
|
||||
|
||||
err = sendStartupAlerts(&c.config, c.startupAlerts)
|
||||
if err == nil && c.expectErr {
|
||||
t.Errorf("sendFirstRunAlerts(%s): Expected error, the code did not error", c.name)
|
||||
} else if err != nil && !c.expectErr {
|
||||
t.Errorf("sendFirstRunAlerts(%s): Did not expect an error, but we got one anyway: %v", c.name, err)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
25
manifest.tmpl
Normal file
25
manifest.tmpl
Normal file
@ -0,0 +1,25 @@
|
||||
image: iamthefij/minitor-go:{{#if build.tag}}{{trimPrefix "v" build.tag}}{{else}}latest{{/if}}
|
||||
{{#if build.tags}}
|
||||
tags:
|
||||
{{#each build.tags}}
|
||||
- {{this}}
|
||||
{{/each}}
|
||||
{{/if}}
|
||||
manifests:
|
||||
-
|
||||
image: iamthefij/minitor-go:{{#if build.tag}}{{trimPrefix "v" build.tag}}-{{/if}}linux-amd64
|
||||
platform:
|
||||
architecture: amd64
|
||||
os: linux
|
||||
-
|
||||
image: iamthefij/minitor-go:{{#if build.tag}}{{trimPrefix "v" build.tag}}-{{/if}}linux-arm64
|
||||
platform:
|
||||
architecture: arm64
|
||||
os: linux
|
||||
variant: v8
|
||||
-
|
||||
image: iamthefij/minitor-go:{{#if build.tag}}{{trimPrefix "v" build.tag}}-{{/if}}linux-arm
|
||||
platform:
|
||||
architecture: arm
|
||||
os: linux
|
||||
variant: v7
|
@ -11,9 +11,9 @@ import (
|
||||
// Monitor represents a particular periodic check of a command
|
||||
type Monitor struct { //nolint:maligned
|
||||
// Config values
|
||||
AlertAfter int16 `yaml:"alert_after"`
|
||||
AlertEvery *int16 `yaml:"alert_every"`
|
||||
CheckInterval SecondsOrDuration `yaml:"check_interval"`
|
||||
AlertAfter int16 `yaml:"alert_after"`
|
||||
AlertEvery *int16 `yaml:"alert_every"`
|
||||
CheckInterval time.Duration `yaml:"check_interval"`
|
||||
Name string
|
||||
AlertDown []string `yaml:"alert_down"`
|
||||
AlertUp []string `yaml:"alert_up"`
|
||||
@ -45,7 +45,7 @@ func (monitor Monitor) ShouldCheck() bool {
|
||||
|
||||
sinceLastCheck := time.Since(monitor.lastCheck)
|
||||
|
||||
return sinceLastCheck >= monitor.CheckInterval.Value()
|
||||
return sinceLastCheck >= monitor.CheckInterval
|
||||
}
|
||||
|
||||
// Check will run the command configured by the Monitor and return a status
|
||||
|
@ -45,9 +45,9 @@ func TestMonitorShouldCheck(t *testing.T) {
|
||||
name string
|
||||
}{
|
||||
{Monitor{}, true, "Empty"},
|
||||
{Monitor{lastCheck: timeNow, CheckInterval: SecondsOrDuration{time.Second * 15}}, false, "Just checked"},
|
||||
{Monitor{lastCheck: timeTenSecAgo, CheckInterval: SecondsOrDuration{time.Second * 15}}, false, "-10s"},
|
||||
{Monitor{lastCheck: timeTwentySecAgo, CheckInterval: SecondsOrDuration{time.Second * 15}}, true, "-20s"},
|
||||
{Monitor{lastCheck: timeNow, CheckInterval: time.Second * 15}, false, "Just checked"},
|
||||
{Monitor{lastCheck: timeTenSecAgo, CheckInterval: time.Second * 15}, false, "-10s"},
|
||||
{Monitor{lastCheck: timeTwentySecAgo, CheckInterval: time.Second * 15}, true, "-20s"},
|
||||
}
|
||||
|
||||
for _, c := range cases {
|
||||
|
@ -1,5 +1,5 @@
|
||||
---
|
||||
check_interval: 1
|
||||
check_interval: 1s
|
||||
|
||||
monitors:
|
||||
- name: Command
|
||||
|
Loading…
Reference in New Issue
Block a user