0.0
No commit activity in last 3 years
No release in over 3 years
sys_watchdog keeps your *NIX servers green by performing periodic checks and optionally actions like service restarts and notifications. https://github.com/tomlobato/sys_watchdog http://sys-watchdog.bettercall.io.
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
 Dependencies

Runtime

~> 2.5
 Project Readme

sys_watchdog

Gem Version Code Climate

By Bettercall.io.

sys_watchdog keeps your *NIX servers green by performing periodic checks and optionally actions like service restarts and notifications.

Take 20min to install and start to think what you`ll make with your spare time ;)

Install

Requirements: Ruby 2.1+ (install on Centos Debian Ubuntu)

Perform this and following steps logged as root user.

gem install sys_watchdog

If using Rbenv, run rbenv rehash to make sys_watchdog binary available.

Setup

sys_watchdog setup

Edit /etc/sys_watchdog.yml. You can see example configurations in util/sys_watchdog_sample.yml and test/sys_watchdog_test.yml.

Test run

Run from command line:

sys_watchdog test

It will execute all system tests defined in /etc/sys_watchdog.yml and exit. You can use this to adjust your tests and get a grasp of the sys_watchdog operation.

Start

Finally, start the periodic run...

sys_watchdog start

Command line Cli

# sys_watchdog help
sys_watchdog v0.1.13

Usage: sys_watchdog [setup|test|start|stop|once|uninstall|help|version]

When called without options executes the system tests each 60 seconds. This is the normal operation when working with a system daemon starter like systemd or upstart.

Options:
  setup        Create the configuration file, create the working directory and configure periodic run with systemd or cron.
  test         Similar to once, but also sets the log output to STDOUT.
  start        Start periodic run. If using systemd it will run 'systemctl start sys_watchdog'. If using cron it will enable/uncomment the cronjob line.  
  stop         Stop periodic run.
  once         Run tests once and exit, rather than run each 60 seconds which is the normal behavior.
  uninstall    Deletes created files and services. After that, if you want to completlly uninstall, run 'gem uninstall sys_watchdog'.
  -v, version  Prints version and exit.
  -h, help     Prints the current message.

Config Settings

config:section of /etc/sys_watchdog.yml:

setting description
name -
server_name -
slack_token -
slack_channel -
smtp_server -
smtp_domain -
mail_from -
mail_to -

Sys Test Settings

tests:section of /etc/sys_watchdog.yml:

setting description
test_cmd -
test_url -
notify_on_output_change -
restore_cmd -
expected_regex -
expected_string -
expected_max -
expected_min -

Minimal /etc/sys_watchdog.yml

config:
  name: My Website Server

  smtp_server: smtp.mymailer.com
  smtp_domain: mysite.com
  mail_from: system@mysite.com
  mail_to: 
    - myself@mysite.com

tests:
  site_status:
    test_url: https://www.mysite.com

  disk_root:
    test_cmd: "df / | grep -v Filesystem | awk '{print $5}' | sed 's/%//'"
    expected_max: 90
    
  worker_status:
    test_cmd: ps aux | grep 'sidekiq 4.2.10 site'

Full sample /etc/sys_watchdog.yml

config:
  name: My Website Server
  server_name: main website # if not present, `hostname` will be used

  # Send alerts to Slack (create your slack token in https://github.com/slack-ruby/slack-ruby-client#usage)
  # slack_token: xoxb-...
  # slack_channel: '#alerts'

  # Send email alerts (via any smtp server: https://sendgrid.com, https://postmarkapp.com/ or your own)
  # smtp_server: smtp.mysite.com
  # smtp_domain: mysite.com
  # mail_from: system@mysite.com
  # mail_to: 
  #   - myself@mysite.com

tests:
  ## General
  boot_time:
    notify_on_output_change: uptime -s

  ## URLs
  site_status:
    test_url: https://www.mysite.com

  # helpdesk_status:
  #   test_url: https://helpdesk_status.mysite.com

  ## Disks
  disk_root:
    test_cmd: "df / | grep -v Filesystem | awk '{print $5}' | sed 's/%//'"
    expected_max: 90

  # disk_shared_storage:
  #   test_cmd: "df /var/www/site/shared | grep -v Filesystem | awk '{print $5}' | sed 's/%//'"
  #   expected_max: 90

  ## Mounts
  # shared_storage:
  #   test_cmd: mount | grep /var/www/site/shared

  ## Services
  # dns_server:
  #   test_cmd: host -t A www.mysite.com ns1.mydns.com | tail -1
  #   expected_regex: ^www.mysite.com has address
  #   restore_cmd: systemctl restart bind9

  # redis_server:
  #   test_cmd: systemctl status redis-server
  #   expected_string: Active: active (running)
  #   restore_cmd: systemctl restart redis-server

  ## Processes
  # worker_status:
  #   test_cmd: ps aux | grep 'sidekiq 4.2.10 site'

Create a Slack Token

From https://github.com/slack-ruby/slack-ruby-client ...

This is something done in Slack, under integrations. Create a new bot, and note its API token.

Develop

Don`t need root privileges here.

Download & install dependencies:

git clone https://github.com/tomlobato/sys_watchdog
cd sys_watchdog
bundle

Run code tests:

chmod 0600 test/sys_watchdog_test.yml 
rake test

Run sys_watchdog tests:

ruby -Ilib/ bin/sys_watchdog test

Send a PR and win a merge! Its free ;)

Copyright and License

Copyright (c) 2017, Bettercall.io.

This project is licensed under the MIT License.