health_check

.gitignore

@@ -0,0 +1,6 @@
+.bundle/
+log/*.log
+pkg/
+test/dummy/db/*.sqlite3
+test/dummy/log/*.log
+test/dummy/tmp/

Gemfile

@@ -0,0 +1,3 @@
+source "https://rubygems.org"
+
+gemspec

MIT-LICENSE

@@ -0,0 +1,3 @@
+MIT License
+
+Copyright (c) Ruling Digital

README.rdoc

@@ -0,0 +1,3 @@
+= HealthCheck
+
+Health check module for orbit-kernel.

Rakefile

@@ -0,0 +1,7 @@
+begin
+  require "bundler/setup"
+rescue LoadError
+  puts "You must `gem install bundler` and `bundle install` to run rake tasks"
+end
+
+Bundler::GemHelper.install_tasks

app/controllers/admin/health_check_monitors_controller.rb

@@ -0,0 +1,79 @@
+class Admin::HealthCheckMonitorsController < OrbitAdminController
+  before_action ->(module_app = @app_title) { set_variables module_app }
+  before_action :set_monitor, only: [:show, :edit, :update, :destroy, :manual_check, :logs, :clean_logs]
+
+  def index
+    @monitors = HealthCheckMonitor.all.desc(:created_at).page(params[:page]).per(10)
+    render :partial => "index" if request.xhr?
+  end
+
+  def show
+    redirect_to logs_admin_health_check_monitor_path(@monitor)
+  end
+
+  def new
+    session[:return_to] = request.referer
+    @monitor = HealthCheckMonitor.new
+  end
+
+  def create
+    @monitor = HealthCheckMonitor.new(monitor_params)
+    @monitor.create_user_id = current_user.id
+    @monitor.update_user_id = current_user.id
+    @monitor.save
+    redirect_to session.delete(:return_to) || admin_health_check_monitors_path
+  end
+
+  def edit
+    session[:return_to] = request.referer
+  end
+
+  def update
+    @monitor.update_attributes(monitor_params)
+    @monitor.update_user_id = current_user.id
+    @monitor.save
+    redirect_to session.delete(:return_to) || admin_health_check_monitors_path
+  end
+
+  def destroy
+    @monitor.destroy
+    redirect_to admin_health_check_monitors_path
+  end
+
+  def manual_check
+    @results = @monitor.check_all
+    alert_logs = []
+    @results.each do |url, result|
+      HealthCheckLog.create(
+        :health_check_monitor_id => @monitor.id,
+        :url => url,
+        :status => result[:status],
+        :response_time => result[:response_time],
+        :error_message => result[:error]
+      )
+      target = @monitor.alert_target(url)
+      alert_logs << target if target
+    end
+    @monitor.send_alert_email(alert_logs) if alert_logs.present?
+    redirect_to logs_admin_health_check_monitor_path(@monitor)
+  end
+
+  def logs
+    @logs = HealthCheckLog.where(:health_check_monitor_id => @monitor.id).desc(:created_at).page(params[:page]).per(20)
+  end
+
+  def clean_logs
+    HealthCheckLog.where(:health_check_monitor_id => @monitor.id).destroy_all
+    redirect_to logs_admin_health_check_monitor_path(@monitor)
+  end
+
+  private
+
+  def set_monitor
+    @monitor = HealthCheckMonitor.find(params[:id])
+  end
+
+  def monitor_params
+    params.require(:health_check_monitor).permit!
+  end
+end

app/helpers/admin/health_check_monitors_helper.rb

@@ -0,0 +1,2 @@
+module Admin::HealthCheckMonitorsHelper
+end

app/models/health_check_alert.rb

@@ -0,0 +1,10 @@
+# encoding: utf-8
+class HealthCheckAlert
+  include Mongoid::Document
+  include Mongoid::Timestamps
+
+  field :health_check_monitor_id, type: BSON::ObjectId
+  field :url, type: String
+
+  index({ health_check_monitor_id: 1, url: 1, created_at: -1 })
+end

app/models/health_check_log.rb

@@ -0,0 +1,14 @@
+# encoding: utf-8
+class HealthCheckLog
+  include Mongoid::Document
+  include Mongoid::Timestamps
+
+  field :url,           type: String
+  field :status,        type: Integer
+  field :response_time, type: Float
+  field :error_message, type: String
+
+  belongs_to :health_check_monitor, class_name: "HealthCheckMonitor", optional: true
+
+  index({ health_check_monitor_id: 1, created_at: -1 })
+end

app/models/health_check_monitor.rb

@@ -0,0 +1,140 @@
+# encoding: utf-8
+require 'net/http'
+require 'uri'
+
+class HealthCheckMonitor
+  include Mongoid::Document
+  include Mongoid::Timestamps
+
+  # 視為失敗的狀態碼。0 代表連不上（逾時、DNS 失敗、憑證錯誤等）
+  # 若只要 502 才算失敗，改成 [502]
+  FAILURE_STATUSES = [0, 500, 502, 503, 504]
+
+  # 告警視窗。只看這段時間內的檢查紀錄，同一網址在這段時間內也只寄一封
+  ALERT_WINDOW = 4.hours
+
+  field :name,                type: String
+  field :target_urls,         type: Array,   default: []
+  field :notification_emails, type: Array,   default: []
+  field :check_interval,      type: Integer, default: 300
+  field :fail_threshold,      type: Integer, default: 3
+  field :timeout,             type: Integer, default: 10
+  field :enabled,             type: Boolean, default: true
+  field :last_checked_at,     type: DateTime
+  field :create_user_id
+  field :update_user_id
+
+  validates :name, presence: true
+
+  scope :is_enabled, -> { where(enabled: true) }
+
+  def target_urls_text
+    (target_urls || []).join("\n")
+  end
+
+  def target_urls_text=(value)
+    self.target_urls = value.to_s.split(/\r?\n/).map(&:strip).reject(&:blank?).map do |u|
+      u.match(/\Ahttps?:\/\//) ? u : "https://#{u}"
+    end
+  end
+
+  def notification_emails_text
+    (notification_emails || []).join(", ")
+  end
+
+  def notification_emails_text=(value)
+    self.notification_emails = value.to_s.split(",").map(&:strip).reject(&:blank?)
+  end
+
+  def threshold
+    fail_threshold.to_i > 0 ? fail_threshold.to_i : 3
+  end
+
+  def check_all
+    results = {}
+    (target_urls || []).each do |url|
+      results[url] = check_url(url)
+    end
+    self.last_checked_at = Time.now
+    self.save
+    results
+  end
+
+  def check_url(url)
+    uri = URI.parse(url)
+    started = Time.now
+    http = Net::HTTP.new(uri.host, uri.port)
+    http.use_ssl = (uri.scheme == 'https')
+    http.verify_mode = OpenSSL::SSL::VERIFY_NONE
+    http.open_timeout = self.timeout
+    http.read_timeout = self.timeout
+    response = http.request(Net::HTTP::Get.new(uri.request_uri))
+    elapsed = ((Time.now - started) * 1000).round(2)
+    { :status => response.code.to_i, :response_time => elapsed, :error => nil }
+  rescue => e
+    { :status => 0, :response_time => 0, :error => e.message }
+  end
+
+  def failure?(status)
+    FAILURE_STATUSES.include?(status.to_i)
+  end
+
+  # 這次檢查寫完 log 之後呼叫。
+  # 條件：4 小時內已檢查滿 threshold 次、最近 threshold 次全部失敗、且 4 小時內還沒寄過這個網址
+  def alert_target(url)
+    n = threshold
+    window_start = Time.now - ALERT_WINDOW
+    logs = HealthCheckLog.where(:health_check_monitor_id => self.id,
+                                :url => url,
+                                :created_at.gte => window_start)
+                         .desc(:created_at).limit(n).to_a
+    return nil if logs.size < n
+    return nil unless logs.all? { |log| failure?(log.status) }
+    return nil if alerted_recently?(url)
+    logs.first
+  end
+
+  def alerted_recently?(url)
+    HealthCheckAlert.where(:health_check_monitor_id => self.id,
+                           :url => url,
+                           :created_at.gte => Time.now - ALERT_WINDOW).exists?
+  end
+
+  def mark_alerted(url)
+    HealthCheckAlert.create(:health_check_monitor_id => self.id, :url => url)
+  end
+
+  # logs 是這次剛好達標的 HealthCheckLog 陣列，全部合併成一封
+  def send_alert_email(logs)
+    return false if notification_emails.blank? || logs.blank?
+    rows = logs.map do |log|
+      {
+        'url'    => log.url.to_s,
+        'status' => log.status.to_s,
+        'error'  => log.error_message.to_s
+      }
+    end
+    email = Email.new(
+      :module_app_key => 'health_check',
+      :mail_to        => notification_emails,
+      :mail_subject   => "[網站監控] #{self.name} - #{logs.size} 個網址失敗",
+      :template       => 'email/health_check_alert.html.erb',
+      :template_data  => {
+        'monitor_name' => self.name.to_s,
+        'threshold'    => threshold,
+        'rows'         => rows,
+        'checked_at'   => logs.first.created_at.strftime('%Y-%m-%d %H:%M:%S')
+      },
+      :mail_sentdate  => Time.now - 1.minute
+    )
+    email.save
+    begin
+      email.deliver
+      logs.each { |log| mark_alerted(log.url) }
+      true
+    rescue => e
+      Rails.logger.error("[health_check] 寄信失敗: #{e}")
+      false
+    end
+  end
+end

app/views/admin/health_check_monitors/_form.html.erb

@@ -0,0 +1,57 @@
+<div class="input-area">
+
+  <div class="control-group">
+    <label class="control-label muted"><%= t("health_check_monitor.name") %></label>
+    <div class="controls">
+      <%= f.text_field :name, :class => "input-block-level" %>
+    </div>
+  </div>
+
+  <div class="control-group">
+    <label class="control-label muted"><%= t("health_check_monitor.target_urls") %></label>
+    <div class="controls">
+      <%= f.text_area :target_urls_text, :rows => 8, :class => "input-block-level", :placeholder => t("health_check_monitor.one_per_line") %>
+    </div>
+  </div>
+
+  <div class="control-group">
+    <label class="control-label muted"><%= t("health_check_monitor.notification_emails") %></label>
+    <div class="controls">
+      <%= f.text_field :notification_emails_text, :class => "input-block-level", :placeholder => t("health_check_monitor.comma_separated") %>
+    </div>
+  </div>
+
+  <div class="control-group">
+    <label class="control-label muted"><%= t("health_check_monitor.check_interval") %></label>
+    <div class="controls">
+      <%= f.number_field :check_interval %>
+    </div>
+  </div>
+
+  <div class="control-group">
+    <label class="control-label muted"><%= t("health_check_monitor.fail_threshold") %></label>
+    <div class="controls">
+      <%= f.number_field :fail_threshold %>
+    </div>
+  </div>
+
+  <div class="control-group">
+    <label class="control-label muted"><%= t("health_check_monitor.timeout") %></label>
+    <div class="controls">
+      <%= f.number_field :timeout %>
+    </div>
+  </div>
+
+  <div class="control-group">
+    <label class="control-label muted"><%= t("health_check_monitor.enabled") %></label>
+    <div class="controls">
+      <%= f.check_box :enabled %>
+    </div>
+  </div>
+
+</div>
+
+<div class="form-actions">
+  <%= f.submit t(:submit), :class => "btn btn-primary" %>
+  <%= link_to t(:cancel), admin_health_check_monitors_path, :class => "btn" %>
+</div>

app/views/admin/health_check_monitors/_index.html.erb

@@ -0,0 +1,39 @@
+<table class="table main-list">
+  <thead>
+    <tr>
+      <th><%= t("health_check_monitor.name") %></th>
+      <th><%= t("health_check_monitor.target_urls") %></th>
+      <th><%= t("health_check_monitor.check_interval") %></th>
+      <th><%= t("health_check_monitor.enabled") %></th>
+      <th><%= t("health_check_monitor.last_checked") %></th>
+    </tr>
+  </thead>
+  <tbody>
+    <% @monitors.each do |monitor| %>
+      <tr>
+        <td>
+          <%= monitor.name %>
+          <div class="quick-edit">
+            <ul class="nav nav-pills">
+              <li><%= link_to t(:edit), edit_admin_health_check_monitor_path(monitor) %></li>
+              <li><%= link_to t("health_check_monitor.logs"), logs_admin_health_check_monitor_path(monitor) %></li>
+              <li><%= link_to t("health_check_monitor.manual_check"), manual_check_admin_health_check_monitor_path(monitor), :method => :post %></li>
+              <li><%= link_to t(:delete_), admin_health_check_monitor_path(monitor), :method => :delete, :data => { :confirm => t(:sure?) } %></li>
+            </ul>
+          </div>
+        </td>
+        <td><%= monitor.target_urls.join("<br/>").html_safe %></td>
+        <td><%= monitor.check_interval %></td>
+        <td><%= monitor.enabled ? t(:yes_) : t(:no_) %></td>
+        <td><%= monitor.last_checked_at.blank? ? "-" : monitor.last_checked_at.strftime("%Y-%m-%d %H:%M") %></td>
+      </tr>
+    <% end %>
+  </tbody>
+</table>
+
+<%=
+  content_tag :div, class: "bottomnav clearfix" do
+    content_tag(:div, paginate(@monitors), class: "pagination pagination-centered") +
+    content_tag(:div, link_to(t(:new_), new_admin_health_check_monitor_path, :class=>"btn btn-primary"), class: "pull-right")
+  end
+%>

app/views/admin/health_check_monitors/edit.html.erb

@@ -0,0 +1,3 @@
+<%= form_for @monitor, :url => admin_health_check_monitor_path(@monitor), :html => { :class => "form-horizontal main-forms" } do |f| %>
+  <%= render :partial => "form", :locals => { :f => f } %>
+<% end %>

app/views/admin/health_check_monitors/index.html.erb

@@ -0,0 +1 @@
+<%= render :partial => "index" %>

app/views/admin/health_check_monitors/logs.html.erb

@@ -0,0 +1,36 @@
+<h3><%= @monitor.name %> - <%= t("health_check_monitor.logs") %></h3>
+
+<p>
+  <%= link_to t("health_check_monitor.manual_check"), manual_check_admin_health_check_monitor_path(@monitor), :method => :post, :class => "btn btn-primary" %>
+  <%= link_to t("health_check_monitor.clean_logs"), clean_logs_admin_health_check_monitor_path(@monitor), :method => :post, :class => "btn", :data => { :confirm => t(:sure?) } %>
+  <%= link_to t(:back), admin_health_check_monitors_path, :class => "btn" %>
+</p>
+
+<table class="table main-list">
+  <thead>
+    <tr>
+      <th>URL</th>
+      <th><%= t("health_check_monitor.status") %></th>
+      <th><%= t("health_check_monitor.response_time") %></th>
+      <th>Error</th>
+      <th><%= t("health_check_monitor.last_checked") %></th>
+    </tr>
+  </thead>
+  <tbody>
+    <% @logs.each do |log| %>
+      <tr>
+        <td><%= log.url %></td>
+        <td><%= log.status %></td>
+        <td><%= log.response_time %></td>
+        <td><%= log.error_message %></td>
+        <td><%= log.created_at.strftime("%Y-%m-%d %H:%M:%S") %></td>
+      </tr>
+    <% end %>
+  </tbody>
+</table>
+
+<div class="bottomnav clearfix">
+  <div class="pagination pagination-centered">
+    <%= paginate @logs %>
+  </div>
+</div>

app/views/admin/health_check_monitors/new.html.erb

@@ -0,0 +1,3 @@
+<%= form_for @monitor, :url => admin_health_check_monitors_path, :html => { :class => "form-horizontal main-forms" } do |f| %>
+  <%= render :partial => "form", :locals => { :f => f } %>
+<% end %>

app/views/email/health_check_alert.html.erb

@@ -0,0 +1,25 @@
+<!DOCTYPE html>
+<html>
+  <head>
+    <meta content='text/html; charset=UTF-8' http-equiv='Content-Type' />
+  </head>
+  <body>
+    <p>監控項目：<%= @data['monitor_name'] %></p>
+    <p>以下 <%= @data['rows'].size %> 個網址連續 <%= @data['threshold'] %> 次檢查失敗：</p>
+    <table border="1" cellpadding="6" style="border-collapse:collapse">
+      <tr>
+        <th>網址</th>
+        <th>狀態碼</th>
+        <th>錯誤訊息</th>
+      </tr>
+      <% @data['rows'].each do |row| %>
+        <tr>
+          <td><%= row['url'] %></td>
+          <td><%= row['status'] %></td>
+          <td><%= row['error'] %></td>
+        </tr>
+      <% end %>
+    </table>
+    <p>檢查時間：<%= @data['checked_at'] %></p>
+  </body>
+</html>

config/locales/en.yml

@@ -0,0 +1,18 @@
+en:
+  health_check: Health Check
+  health_check_monitor:
+    name: Name
+    target_urls: Target URLs
+    check_interval: Check Interval (sec)
+    fail_threshold: Alert After N Consecutive Failures
+    timeout: Timeout (sec)
+    notification_emails: Notification Emails
+    enabled: Enabled
+    status: Status
+    response_time: Response Time (ms)
+    last_checked: Last Checked
+    manual_check: Check Now
+    logs: Logs
+    clean_logs: Clear All Logs
+    one_per_line: One URL per line, must include http:// or https:// (defaults to https://)
+    comma_separated: Comma separated

config/locales/zh_tw.yml

@@ -0,0 +1,18 @@
+zh_tw:
+  health_check: 健康檢查
+  health_check_monitor:
+    name: 監控名稱
+    target_urls: 監控網址
+    check_interval: 檢查間隔(秒)
+    fail_threshold: 連續失敗幾次寄信
+    timeout: 逾時(秒)
+    notification_emails: 通知信箱
+    enabled: 啟用
+    status: 狀態
+    response_time: 回應時間(ms)
+    last_checked: 最後檢查
+    manual_check: 立即檢查
+    logs: 檢查紀錄
+    clean_logs: 清除全部紀錄
+    one_per_line: 一行一個網址，需含 http:// 或 https://（未填會自動補 https://）
+    comma_separated: 以逗號分隔

config/routes.rb

@@ -0,0 +1,17 @@
+Rails.application.routes.draw do
+
+locales = Site.find_by(site_active: true).in_use_locales rescue I18n.available_locales
+
+  scope "(:locale)", locale: Regexp.new(locales.join("|")) do
+    namespace :admin do
+      resources :health_check_monitors do
+        member do
+          post :manual_check
+          get :logs
+          post :clean_logs
+        end
+      end
+    end
+  end
+
+end

health_check.gemspec

@@ -0,0 +1,18 @@
+$:.push File.expand_path("../lib", __FILE__)
+
+# Maintain your gem's version:
+require "health_check/version"
+
+# Describe your gem and declare its dependencies:
+Gem::Specification.new do |s|
+  s.name        = "health_check"
+  s.version     = HealthCheck::VERSION
+  s.authors     = ["Orbitek"]
+  s.email       = ["service@orbitek.co"]
+  s.homepage    = "https://orbitek.co"
+  s.summary     = "Health Check for orbitek"
+  s.description = "Health Check for orbitek"
+  s.license     = "MIT"
+
+  s.files = Dir["{app,config,db,lib}/**/*", "MIT-LICENSE", "Rakefile", "README.rdoc"]
+end

lib/health_check.rb

@@ -0,0 +1,4 @@
+require "health_check/engine"
+
+module HealthCheck
+end

lib/health_check/engine.rb

@@ -0,0 +1,33 @@
+module HealthCheck
+  class Engine < ::Rails::Engine
+    initializer "health_check" do
+      Rails.application.config.to_prepare do
+        OrbitApp.registration "HealthCheck", :type => "ModuleApp" do
+          module_label "health_check.health_check"
+          base_url File.expand_path File.dirname(__FILE__)
+
+          authorizable
+
+          side_bar do
+            head_label_i18n 'health_check', :icon_class=>"icons-cycle"
+            available_for "users"
+            active_for_controllers (['admin/health_check_monitors'])
+            head_link_path "admin_health_check_monitors_path"
+
+            context_link 'list_',
+                                :link_path=>"admin_health_check_monitors_path" ,
+                                :priority=>1,
+                                :active_for_action=>{'admin/health_check_monitors'=>'index'},
+                                :available_for => 'users'
+
+            context_link 'new_',
+                                :link_path=>"new_admin_health_check_monitor_path" ,
+                                :priority=>2,
+                                :active_for_action=>{'admin/health_check_monitors'=>'new'},
+                                :available_for => 'sub_managers'
+          end
+        end
+      end
+    end
+  end
+end

lib/health_check/version.rb

@@ -0,0 +1,3 @@
+module HealthCheck
+  VERSION = "0.0.1"
+end

lib/tasks/health_check_tasks.rake

@@ -0,0 +1,3 @@
+# desc "Explaining what the task does"
+# task :health_check do
+# end