rud

Chunked event publishing - design overview

Class: EventCollector

Roles:

• #enqueue_event(payload, target queue name): stores messages to be sent in a simple array
• #publish() send everything currently enqueued to external queue system (here: RabbitMQ); if the publish fails for any reason, messages are kept in memory for later retry (along with grumbling in logs, etc).
• #clear_events_to_publish! is useful in rollback scenarios
• .thread_singleton_instance accessor with lazy initialization, like ActiveRecord::Base.connection. This allows for easy setup, and persistent state across requests (should flushing fail we can retry on next request on the same thread). The thread_singleton_instance also easy replacement with a simple test implementation, think ActionMailer. Sending actual messages in test is best restricted to a few integration test scenarios.

rud

# Do not suffer slow queries in production:
aggressive_timeout_settings: &aggressive_timeout_settings
  adapter: postgresql
  encoding: unicode
  connect_timeout:  1 # second - network handshake time
  checkout_timeout: 1 # second - connection pool checkout wait time
  variables:
    statement_timeout: 2500 # ms - maximum time for a single SQL query

production:

rud

### Keybase proof

I hereby claim:

  * I am rud on github.
  * I am rud (https://keybase.io/rud) on keybase.
  * I have a public key whose fingerprint is 2949 8980 F3BE 3D8F FC0B  E75D 3E50 71F8 6896 12B4

To claim this, I am signing this object:

rud

#!/bin/sh

# Simple script for pruning local branches already merged,
# and remove remote tracking branches that have been deleted

# Default to "origin" if no remote is passed
remote=${1-origin}

# Branches to keep, regardless of their current merge state
permanent_branches="develop|master|production"

rud

# --------------------------------------------------------
# vol as a shell script
# --------------------------------------------------------
# You would put this code in a directory that's in your
# path, like ~/bin (PATH=~/bin:$PATH in your .bashrc to
# add a directory like that to your path.)
# The invokation syntax is the same as that of the
# function.
# --------------------------------------------------------
# Example usage:

rud

# --------------------------------------------------------
# vol as a shell script
# --------------------------------------------------------
# You would put this code in a directory that's in your
# path, like ~/bin (PATH=~/bin:$PATH in your .bashrc to
# add a directory like that to your path.)
# The invokation syntax is the same as that of the
# function.
# --------------------------------------------------------
# Example usage:

rud

class ToggleLogging
  def self.setup
    Signal.trap('USR1') do
      toggle
    end
  end

  def self.toggle
    level_before = Rails.logger.level == Logger::DEBUG
    Rails.logger.level = debug_before ? Logger::INFO : Logger::DEBUG

rud

If you use tests of the form: User.accessible_attributes.should =~ [:email, :password] you'll get an extra attribute called "" if you set config.active_record.whitelist_attributes = true, and thus a specfailure. Annoying.

Instead, you can add this to your config/application.rb:

# Require attr_accessible to be used on all models for mass-assignment to work
ActiveRecord::Base.instance_eval do
  attr_accessible # none - and this method doesn't whitelist an attribute with the name ""
end

rud

PATH
  remote: .
  specs:
    em-postgresql-adapter (0.2)
      activerecord (>= 3.1.0)
      eventmachine
      pg (>= 0.8.0)

GEM
  remote: http://rubygems.org/

rud

listen 3000

worker_processes 1
timeout          30

preload_app true


after_fork do |server, worker|
  require "amqp"