Fix ownership of app directory

This seems to be causing an error on startup.

.drone.yml

@@ -1,13 +1,59 @@
+---
 kind: pipeline
 name: test
 
 steps:
-  - name: build
-    image: golang:1.12
-    commands:
-      - make build
 
   - name: test
     image: golang:1.12
     commands:
+      - make build
       - make test
+
+  - name: check
+    image: python:3
+    commands:
+      - pip install pre-commit==1.20.0
+      - make check
+
+  - name: notify
+    image: drillster/drone-email
+    settings:
+      host:
+        from_secret: SMTP_HOST
+      username:
+        from_secret: SMTP_USER
+      password:
+        from_secret: SMTP_PASS
+      from: drone@iamthefij.com
+    when:
+      status: [changed, failure]
+
+---
+kind: pipeline
+name: publish
+
+depends_on:
+  - test
+
+trigger:
+  event:
+    - push
+    - tag
+  refs:
+    - refs/heads/master
+    - refs/tags/v*
+
+steps:
+
+  # Might consider moving this step into the previous pipeline
+  - name: push image
+    image: plugins/docker
+    settings:
+      repo: iamthefij/minitor-go
+      dockerfile: Dockerfile.multi-stage
+      auto_tag: true
+      username:
+        from_secret: docker_username
+      password:
+        from_secret: docker_password

.pre-commit-config.yaml

@@ -0,0 +1,19 @@
+---
+repos:
+  - repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v2.4.0
+    hooks:
+      - id: check-added-large-files
+      - id: check-yaml
+        args:
+          - --allow-multiple-documents
+      - id: trailing-whitespace
+      - id: end-of-file-fixer
+      - id: check-merge-conflict
+  - repo: git://github.com/dnephin/pre-commit-golang
+    rev: v0.3.5
+    hooks:
+      - id: go-fmt
+      - id: go-imports
+      # - id: gometalinter
+      # - id: golangci-lint

Dockerfile

@@ -1,8 +1,24 @@
 ARG REPO=library
-FROM ${REPO}/busybox:latest
-WORKDIR /root/
+FROM ${REPO}/alpine:3.10
+RUN mkdir /app
+WORKDIR /app/
 
+# Copy minitor in
 ARG ARCH=amd64
 COPY ./minitor-go ./minitor
 
+# Add common checking tools
+RUN apk --no-cache add bash=~5.0 curl=~7.66 jq=~1.6
+
+# Add minitor user for running as non-root
+RUN addgroup -S minitor && adduser -S minitor -G minitor
+
+# Copy scripts
+COPY ./scripts /app/scripts
+RUN chown -R minitor:minitor /app
+RUN chmod -R 755 /app/scripts
+
+# Drop to non-root user
+USER minitor
+
 ENTRYPOINT [ "./minitor" ]

Dockerfile.multi-stage

@@ -1,7 +1,7 @@
 ARG REPO=library
 FROM golang:1.12-alpine AS builder
 
-RUN apk add --no-cache git
+RUN apk add --no-cache git=~2
 
 RUN mkdir /app
 WORKDIR /app
@@ -16,8 +16,27 @@ ARG VERSION=dev
 ENV CGO_ENABLED=0 GOOS=linux GOARCH=${ARCH}
 RUN go build -ldflags "-X main.version=${VERSION}" -a -installsuffix nocgo -o minitor .
 
-FROM ${REPO}/busybox:latest
-WORKDIR /root/
+FROM ${REPO}/alpine:3.10
+RUN mkdir /app
+WORKDIR /app/
+
+# Copy minitor in
 COPY --from=builder /app/minitor .
 
+# Add common checking tools
+RUN apk --no-cache add bash=~5.0 curl=~7.66 jq=~1.6
+
+# Add minitor user for running as non-root
+RUN addgroup -S minitor && adduser -S minitor -G minitor
+
+# Copy scripts
+COPY ./scripts /app/scripts
+RUN chown -R minitor:minitor /app
+RUN chmod -R 755 /app/scripts
+
+# Drop to non-root user
+USER minitor
+
 ENTRYPOINT [ "./minitor" ]
+
+# vim: set filetype=dockerfile:

Makefile

@@ -1,6 +1,7 @@
+.PHONY: all
 DOCKER_TAG ?= minitor-go-${USER}
 
-.PHONY: test
+.PHONY: default
 default: test
 
 .PHONY: build
@@ -14,6 +15,10 @@ minitor-go:
 run: minitor-go build
 	./minitor-go -debug
 
+.PHONY: run-metrics
+run-metrics: minitor-go build
+	./minitor-go -debug -metrics
+
 .PHONY: test
 test:
 	go test -coverprofile=coverage.out
@@ -24,6 +29,15 @@ test:
 	@go tool cover -func=coverage.out | awk -v target=80.0% \
 		'/^total:/ { print "Total coverage: " $$3 " Minimum coverage: " target; if ($$3+0.0 >= target+0.0) print "ok"; else { print "fail"; exit 1; } }'
 
+# Installs pre-commit hooks
+.PHONY: install-hooks
+install-hooks:
+	pre-commit install --install-hooks
+
+# Checks files for encryption
+.PHONY: check
+check:
+	pre-commit run --all-files
 
 .PHONY: clean
 clean:

README.md

@@ -2,7 +2,7 @@
 
 A reimplementation of [Minitor](https://git.iamthefij/iamthefij/minitor) in Go
 
-Minitor is already a very minimal monitoring tool. Python 3 was a quick way to get something live, but Python itself comes with a very large footprint.Thus Go feels like a better fit for the project, longer term.
+Minitor is already a minimal monitoring tool. Python 3 was a quick way to get something live, but Python itself comes with a large footprint. Thus Go feels like a better fit for the project, longer term.
 
 Initial target is meant to be roughly compatible requiring only minor changes to configuration. Future iterations may diverge to take advantage of Go specific features.
 
@@ -30,7 +30,7 @@ monitors:
     command_shell: echo 'test'
 ```
 
-Second, templating for Alert messages has been updated. In the Python version, `str.format(...)` was used with certain keys passed in that could be used to format messages. In the Go version, we use a struct containing Alert info and the built in Go templating format. Eg.
+Second, templating for Alert messages has been updated. In the Python version, `str.format(...)` was used with certain keys passed in that could be used to format messages. In the Go version, we use a struct, `AlertNotice` defined in `alert.go` and the built in Go templating format. Eg.
 
 minitor-py:
 ```yaml
@@ -38,7 +38,7 @@ alerts:
   log_command:
     command: ['echo', '{monitor_name}']
   log_shell:
-    command_shell: "echo {monitor_name}"
+    command_shell: 'echo {monitor_name}'
 ```
 
 minitor-go:
@@ -47,7 +47,7 @@ alerts:
   log_command:
     command: ['echo', '{{.MonitorName}}']
   log_shell:
-    command_shell: "echo {{.MonitorName}}"
+    command_shell: 'echo {{.MonitorName}}'
 ```
 
 Finally, newlines in a shell command don't terminate a particular command. Semicolons must be used and continuations should not.
@@ -84,10 +84,11 @@ Pairity:
   - [x] Run alert commands
   - [x] Run alert commands in a shell
   - [x] Allow templating of alert commands
-  - [ ] Implement Prometheus client to export metrics
-  - [ ] Test coverage
+  - [x] Implement Prometheus client to export metrics
+  - [x] Test coverage
+  - [ ] Integration testing (manual or otherwise)
 
-Improvement:
+Improvement (potentially breaking):
 
   - [ ] Implement leveled logging (maybe glog or logrus)
   - [ ] Consider switching from YAML to TOML
@@ -95,3 +96,4 @@ Improvement:
   - [ ] Consider dropping `alert_up` and `alert_down` in favor of using Go templates that offer more control of messaging
   - [ ] Async checking
   - [ ] Use durations rather than seconds checked in event loop
+  - [ ] Revisit metrics and see if they all make sense

config.go

@@ -2,10 +2,11 @@ package main
 
 import (
 	"errors"
-	"gopkg.in/yaml.v2"
 	"io/ioutil"
 	"log"
 	"os"
+
+	"gopkg.in/yaml.v2"
 )
 
 // Config type is contains all provided user configuration
@@ -84,7 +85,9 @@ func LoadConfig(filePath string) (config Config, err error) {
 		return
 	}
 
-	log.Printf("config:\n%v\n", config)
+	if LogDebug {
+		log.Printf("DEBUG: Config values:\n%v\n", config)
+	}
 
 	if !config.IsValid() {
 		err = errors.New("Invalid configuration")

main.go

@@ -3,9 +3,7 @@ package main
 import (
 	"flag"
 	"fmt"
-	"github.com/prometheus/client_golang/prometheus/promhttp"
 	"log"
-	"net/http"
 	"time"
 )
 
@@ -15,6 +13,10 @@ var (
 
 	// ExportMetrics will track whether or not we want to export metrics to prometheus
 	ExportMetrics = false
+	// MetricsPort is the port to expose metrics on
+	MetricsPort = 8080
+	// Metrics contains all active metrics
+	Metrics = NewMetrics()
 
 	// version of minitor being run
 	version = "dev"
@@ -23,7 +25,13 @@ var (
 func checkMonitors(config *Config) error {
 	for _, monitor := range config.Monitors {
 		if monitor.ShouldCheck() {
-			_, alertNotice := monitor.Check()
+			success, alertNotice := monitor.Check()
+
+			hasAlert := alertNotice != nil
+
+			// Track status metrics
+			Metrics.SetMonitorStatus(monitor.Name, success)
+			Metrics.CountCheck(monitor.Name, success, hasAlert)
 
 			// Should probably consider refactoring everything below here
 			if alertNotice != nil {
@@ -55,6 +63,9 @@ func checkMonitors(config *Config) error {
 								err,
 							)
 						}
+
+						// Count alert metrics
+						Metrics.CountAlert(monitor.Name, alert.Name)
 					} else {
 						// This case should never actually happen since we validate against it
 						log.Printf("ERROR: Unknown alert for monitor %s: %s", alertNotice.MonitorName, alertName)
@@ -68,11 +79,6 @@ func checkMonitors(config *Config) error {
 	return nil
 }
 
-func serveMetrics() {
-	http.Handle("/metrics", promhttp.Handler())
-	_ = http.ListenAndServe(":8080", nil)
-}
-
 func main() {
 	// Get debug flag
 	flag.BoolVar(&LogDebug, "debug", false, "Enables debug logs (default: false)")
@@ -95,7 +101,7 @@ func main() {
 	// Serve metrics exporter, if specified
 	if ExportMetrics {
 		log.Println("INFO: Exporting metrics to Prometheus")
-		go serveMetrics()
+		go ServeMetrics()
 	}
 
 	// Start main loop

metrics.go

@@ -0,0 +1,101 @@
+package main
+
+import (
+	"fmt"
+	"net/http"
+
+	"github.com/prometheus/client_golang/prometheus"
+	"github.com/prometheus/client_golang/prometheus/promhttp"
+)
+
+// TODO: Not sure if this is the best way to handle. A global instance for
+// metrics isn't bad, but it might be nice to curry versions of the metrics
+// for each monitor. Especially since every monitor has it's own. Perhaps
+// another new function that essentially curries each metric for a given
+// monitor name would do. This could be run when validating monitors and
+// initializing alert templates.
+
+// MinitorMetrics contains all counters and metrics that Minitor will need to access
+type MinitorMetrics struct {
+	alertCount    *prometheus.CounterVec
+	checkCount    *prometheus.CounterVec
+	monitorStatus *prometheus.GaugeVec
+}
+
+// NewMetrics creates and initializes all metrics
+func NewMetrics() *MinitorMetrics {
+	// Initialize all metrics
+	metrics := &MinitorMetrics{
+		alertCount: prometheus.NewCounterVec(
+			prometheus.CounterOpts{
+				Name: "minitor_alert_total",
+				Help: "Number of Minitor alerts",
+			},
+			[]string{"alert", "monitor"},
+		),
+		checkCount: prometheus.NewCounterVec(
+			prometheus.CounterOpts{
+				Name: "minitor_check_total",
+				Help: "Number of Minitor checks",
+			},
+			[]string{"monitor", "status", "is_alert"},
+		),
+		monitorStatus: prometheus.NewGaugeVec(
+			prometheus.GaugeOpts{
+				Name: "minitor_monitor_up_count",
+				Help: "Status of currently responsive monitors",
+			},
+			[]string{"monitor"},
+		),
+	}
+
+	// Register newly created metrics
+	prometheus.MustRegister(metrics.alertCount)
+	prometheus.MustRegister(metrics.checkCount)
+	prometheus.MustRegister(metrics.monitorStatus)
+
+	return metrics
+}
+
+// SetMonitorStatus sets the current status of Monitor
+func (metrics *MinitorMetrics) SetMonitorStatus(monitor string, isUp bool) {
+	val := 0.0
+	if isUp {
+		val = 1.0
+	}
+	metrics.monitorStatus.With(prometheus.Labels{"monitor": monitor}).Set(val)
+}
+
+// CountCheck counts the result of a particular Monitor check
+func (metrics *MinitorMetrics) CountCheck(monitor string, isSuccess bool, isAlert bool) {
+	status := "failure"
+	if isSuccess {
+		status = "success"
+	}
+
+	alertVal := "false"
+	if isAlert {
+		alertVal = "true"
+	}
+
+	metrics.checkCount.With(
+		prometheus.Labels{"monitor": monitor, "status": status, "is_alert": alertVal},
+	).Inc()
+}
+
+// CountAlert counts an alert
+func (metrics *MinitorMetrics) CountAlert(monitor string, alert string) {
+	metrics.alertCount.With(
+		prometheus.Labels{
+			"alert":   alert,
+			"monitor": monitor,
+		},
+	).Inc()
+}
+
+// ServeMetrics starts an http server with a Prometheus metrics handler
+func ServeMetrics() {
+	http.Handle("/metrics", promhttp.Handler())
+	host := fmt.Sprintf(":%d", MetricsPort)
+	_ = http.ListenAndServe(host, nil)
+}

sample-config.yml

@@ -1,29 +1,41 @@
-check_interval: 30
+---
+check_interval: 5
 
 monitors:
-  - name: My Website
+  - name: Fake Website
     command: ['curl', '-s', '-o', '/dev/null', 'https://minitor.mon']
-    alert_down: [ log, mailgun_down, sms_down ]
-    alert_up: [ log, email_up ]
-    check_interval: 30 # Must be at minimum the global `check_interval`
+    alert_down: [log_down, mailgun_down, sms_down]
+    alert_up: [log_up, email_up]
+    check_interval: 10  # Must be at minimum the global `check_interval`
     alert_after: 3
     alert_every: -1  # Defaults to -1 for exponential backoff. 0 to disable repeating
+  - name: Real Website
+    command: ['curl', '-s', '-o', '/dev/null', 'https://google.com']
+    alert_down: [log_down, mailgun_down, sms_down]
+    alert_up: [log_up, email_up]
+    check_interval: 5
+    alert_after: 3
+    alert_every: -1
 
 alerts:
+  log_down:
+    command: ["echo", "Minitor failure for {{.MonitorName}}"]
+  log_up:
+    command: ["echo", "Minitor recovery for {{.MonitorName}}"]
   email_up:
     command: [sendmail, "me@minitor.mon", "Recovered: {monitor_name}", "We're back!"]
   mailgun_down:
-    command: >
+    command_shell: >
       curl -s -X POST
-      -F subject="Alert! {monitor_name} failed"
+      -F subject="Alert! {{.MonitorName}} failed"
       -F from="Minitor <minitor@minitor.mon>"
       -F to=me@minitor.mon
       -F text="Our monitor failed"
       https://api.mailgun.net/v3/minitor.mon/messages
       -u "api:${MAILGUN_API_KEY}"
   sms_down:
-    command: >
-      curl -s -X POST -F "Body=Failure! {monitor_name} has failed"
+    command_shell: >
+      curl -s -X POST -F "Body=Failure! {{.MonitorName}} has failed"
       -F "From=${AVAILABLE_NUMBER}" -F "To=${MY_PHONE}"
       "https://api.twilio.com/2010-04-01/Accounts/${ACCOUNT_SID}/Messages"
       -u "${ACCOUNT_SID}:${AUTH_TOKEN}"

scripts/README.md

@@ -0,0 +1,5 @@
+# Minitor Scripts
+
+A collection of some handy scripts to use with Minitor
+
+These are not included with the Python package, but they are included in the Docker image in `/app/scripts`.

scripts/docker_check.sh

@@ -0,0 +1,51 @@
+#! /bin/bash
+set -e
+
+#################
+# docker_check.sh
+#
+# Checks the most recent state exit code of a Docker container
+#################
+
+# Docker host will default to a socket
+# To override, export DOCKER_HOST to a new hostname
+DOCKER_HOST="${DOCKER_HOST:=socket}"
+container_name="$1"
+
+# Curls Docker either using a socket or URL
+function curl_docker {
+    local path="$1"
+    if [ "$DOCKER_HOST" == "socket" ]; then
+        curl --unix-socket /var/run/docker.sock "http://localhost/$path" 2>/dev/null
+    else
+        curl "http://${DOCKER_HOST}/$path" 2>/dev/null
+    fi
+}
+
+# Returns caintainer ID for a given container name
+function get_container_id {
+    local container_name="$1"
+    curl_docker 'containers/json?all=1' \
+        | jq -r ".[] | {Id, Name: .Names[]} | select(.Name == \"/${container_name}\") | .Id"
+}
+
+# Returns container JSON
+function inspect_container {
+    local container_id=$1
+    curl_docker "containers/$container_id/json"
+}
+
+if [ -z "$container_name" ]; then
+    echo "Usage: $0 container_name"
+    echo "Will exit with the last status code of continer with provided name"
+    exit 1
+fi
+
+container_id=$(get_container_id $container_name)
+if [ -z "$container_id" ]; then
+    echo "ERROR: Could not find container with name: $container_name"
+    exit 1
+fi
+exit_code=$(inspect_container "$container_id" | jq -r .State.ExitCode)
+
+exit "$exit_code"

scripts/docker_healthcheck.sh

@@ -0,0 +1,61 @@
+#! /bin/bash
+set -e
+
+#################
+# docker_healthcheck.sh
+#
+# Returns the results of a Docker Healthcheck for a container
+#################
+
+# Docker host will default to a socket
+# To override, export DOCKER_HOST to a new hostname
+DOCKER_HOST="${DOCKER_HOST:=socket}"
+container_name="$1"
+
+# Curls Docker either using a socket or URL
+function curl_docker {
+    local path="$1"
+    if [ "$DOCKER_HOST" == "socket" ]; then
+        curl --unix-socket /var/run/docker.sock "http://localhost/$path" 2>/dev/null
+    else
+        curl "http://${DOCKER_HOST}/$path" 2>/dev/null
+    fi
+}
+
+# Returns caintainer ID for a given container name
+function get_container_id {
+    local container_name="$1"
+    curl_docker 'containers/json?all=1' \
+        | jq -r ".[] | {Id, Name: .Names[]} | select(.Name == \"/${container_name}\") | .Id"
+}
+
+# Returns container JSON
+function inspect_container {
+    local container_id="$1"
+    curl_docker "containers/$container_id/json"
+}
+
+if [ -z "$container_name" ]; then
+    echo "Usage: $0 container_name"
+    echo "Will return results of healthcheck for continer with provided name"
+    exit 1
+fi
+
+container_id=$(get_container_id "$container_name")
+if [ -z "$container_id" ]; then
+    echo "ERROR: Could not find container with name: $container_name"
+    exit 1
+fi
+health=$(inspect_container "$container_id" | jq -r '.State.Health.Status')
+
+case "$health" in
+    null)
+        echo "No healthcheck results"
+        ;;
+    starting|healthy)
+        echo "Status: '$health'"
+        ;;
+    *)
+        echo "Status: '$health'"
+        exit 1
+esac

test/invalid-config-missing-alerts.yml

@@ -6,4 +6,3 @@ monitors:
     alert_down: [ 'alert_down', 'log_shell', 'log_command' ]
     # alert_every: -1
     alert_every: 0
-