Skip to content

Latest commit

 

History

History
122 lines (95 loc) · 3.58 KB

capistrano.md

File metadata and controls

122 lines (95 loc) · 3.58 KB
Raw

Deploy with Capistrano

These instructions should be good enough to get you started deploying capistrano with Errbit. More than likely, you'll have to adjust some things to suit your needs, so you should understand how to use capistrano before you continue.

Clone and prepare the source code repository

git clone git@github.com:errbit/errbit.git
cd errbit

# Create and edit deploy.rb
cp config/deploy.example.rb config/deploy.rb
$EDITOR config/deploy.rb

# Create and edit production.rb
cp config/deploy/production.example.rb config/deploy/production.rb
$EDITOR config/deploy/production.rb

# Create the configs yourself, or run errbit:setup to upload the
# defaults
bundle exec cap production errbit:setup

# Check to make sure configs exist
bundle exec cap production deploy:check

# Deploy
bundle exec cap production deploy

# Setup the remote DB if you haven't already
bundle exec cap production db:setup

Static Assets

For a deployment of any real size, you'll probably want to set up a web server for efficiently serving static assets. If you choose to go this route, just map all requests for /assets/.* to /deploy/path/shared/public/assets

Starting Errbit

Errbit comes with some capistrano tasks to manage running Errbit under puma. To start Errbit, you can run:

bundle exec cap production puma:start

Status of Errbit

To check if the Errbit server is running you can run:

bundle exec cap production puma:status

Stopping Errbit

To stop Errbit run

bundle exec cap production puma:stop

Supervising and monitoring Errbit is beyond the scope of this documentation.

rbenv support

Pass rbenv environment when running cap to use rbenv. See capistrano/rbenv for more information.

rbenv=1 bundle exec cap production deploy

Schedule recurring tasks

You may want to periodically clear resolved errors to free up space. Schedule rake errbit:clear_resolved to run every day or so.

Monit

If you like errbit to be monitored by monit, you'll have to install and start monit with http support before deploying errbit. In order to enable http support you have to edit the monit config file which you can find in /etc/monit/monitrc for Ubuntu and set these lines:

set httpd port 2812 and
   use address localhost  # only accept connection from localhost
   allow localhost

Next you have to add the following line to the Capfile:

require 'capistrano/puma/monit'

And you have to deploy the monit config with the command:

bundle exec cap production puma:monit:config

The configuration file (depending on the distro) will be generated at: /etc/monit/conf.d/puma_errbit_production.conf

Controlling memory usage with monit

If you like to limit memory usage for puma and restart it in case the usage gets over a 2GB limit, for example, you can add at the end of the monit config file the line

if totalmem is greater than 2048 MB for 3 cycles then restart

The config file will look like:

# Monit configuration for Puma
# Service name: puma_errbit_production
#
check process puma_errbit_production
  with pidfile "/var/www/apps/errbit/shared/tmp/pids/puma.pid"
  start program = "/usr/bin/sudo -u root /bin/bash -c 'cd /var/www/apps/errbit/current && /usr/bin/env bundle exec puma -C /var/www/apps/errbit/shared/puma.rb --daemon'"
  stop program = "/usr/bin/sudo -u root /bin/bash -c 'cd /var/www/apps/errbit/current && /usr/bin/env bundle exec pumactl -S /var/www/apps/errbit/shared/tmp/pids/puma.state stop'"
  if totalmem is greater than 2048 MB for 3 cycles then restart