No longer raise exception when scheduler job fails to run after ten tries, but record the error instead.

app/assets/javascripts/app/controllers/monitoring.coffee

@@ -4,6 +4,7 @@ class Index extends App.ControllerSubContent
   events:
     'click .js-resetToken': 'resetToken'
     'click .js-select': 'selectAll'
+    'click .js-restartDeadJobs': 'restartDeadJobs'
 
   constructor: ->
     super
@@ -29,7 +30,10 @@ class Index extends App.ControllerSubContent
     )
 
   render: =>
-    @html App.view('monitoring')(data: @data)
+    @html App.view('monitoring')(
+      data: @data
+      job_restart_count: @job_restart_count
+    )
 
   resetToken: (e) =>
     e.preventDefault()
@@ -42,4 +46,15 @@ class Index extends App.ControllerSubContent
         @load()
     )
 
+  restartDeadJobs: (e) =>
+    e.preventDefault()
+    @ajax(
+      id:    'restart_dead_jobs_request'
+      type:  'POST'
+      url:   "#{@apiPath}/monitoring/restart_dead_jobs"
+      success: (data) =>
+        @job_restart_count = data.job_restart_count
+        @render()
+    )
+
 App.Config.set('Monitoring', { prio: 3600, name: 'Monitoring', parent: '#system', target: '#system/monitoring', controller: Index, permission: ['admin.monitoring'] }, 'NavBarAdmin')

app/assets/javascripts/app/views/monitoring.jst.eco

@@ -34,6 +34,15 @@
         <% end %>
       <% end %>
     </ul>
+    <% if !_.isEmpty(@data.issues): %>
+      <button class="btn btn--primary js-restartDeadJobs"><%- @T('Restart Dead Jobs') %></button>
+      <% if !_.isUndefined(@job_restart_count): %>
+        <p>
+          <%- @T('Detected %s dead job(s) available for restart', @job_restart_count) %>
+          <%- ', restarting...' if @job_restart_count > 0 %>
+        </p>
+      <% end %>
+    <% end %>
   </div>
 
 </div>

app/controllers/monitoring_controller.rb

@@ -81,6 +81,10 @@ curl http://localhost/api/v1/monitoring/health_check?token=XXX
       issues.push 'scheduler not running'
     end
 
+    Scheduler.where(status: 'error').each { |scheduler|
+      issues.push "Failed to run scheduled job \'#{scheduler.name}\'. Cause: #{scheduler.error_message}"
+    }
+
     token = Setting.get('monitoring_token')
 
     if issues.empty?
@@ -173,6 +177,22 @@ curl http://localhost/api/v1/monitoring/status?token=XXX
     render json: result, status: :created
   end
 
+  def restart_dead_jobs
+    access_check
+
+    count = 0
+    Scheduler.where(status: 'error').where(active: false).each { |scheduler|
+      scheduler.active = true
+      scheduler.save
+      count += 1
+    }
+
+    result = {
+      job_restart_count: count
+    }
+    render json: result
+  end
+
   private
 
   def token_or_permission_check

app/models/scheduler.rb

@@ -171,6 +171,8 @@ class Scheduler < ApplicationModel
   def self._start_job(job, try_count = 0, try_run_time = Time.zone.now)
     job.last_run      = Time.zone.now
     job.pid           = Thread.current.object_id
+    job.status        = 'ok'
+    job.error_message = ''
     job.save
     logger.info "execute #{job.method} (try_count #{try_count})..."
     eval job.method() # rubocop:disable Lint/Eval
@@ -197,7 +199,14 @@ class Scheduler < ApplicationModel
     if try_run_max > try_count
       _start_job(job, try_count, try_run_time)
     else
-      raise "STOP thread for #{job.method} after #{try_count} tries (#{e.inspect})"
+      @@jobs_started[ job.id ] = false
+      error = "Failed to run #{job.method} after #{try_count} tries #{e.inspect}"
+      logger.error error
+
+      job.error_message = error
+      job.status        = 'error'
+      job.active        = false
+      job.save
     end
   end
 

config/routes/monitoring.rb

@@ -4,5 +4,6 @@ Zammad::Application.routes.draw do
   match api_path + '/monitoring/health_check',         to: 'monitoring#health_check',      via: :get
   match api_path + '/monitoring/status',               to: 'monitoring#status',            via: :get
   match api_path + '/monitoring/token',                to: 'monitoring#token',             via: :post
+  match api_path + '/monitoring/restart_dead_jobs',    to: 'monitoring#restart_dead_jobs', via: :post
 
 end

db/migrate/20170515000001_scheduler_status.rb

@@ -0,0 +1,8 @@
+class SchedulerStatus < ActiveRecord::Migration
+  def up
+    change_table :schedulers do |t|
+      t.string :error_message
+      t.string :status
+    end
+  end
+end