RubyFlow : The Ruby Community Blog

Home   Submit   Sign Up   Log In   leaders   Twitter   RSS Feed  

itsderek23 — 21 posts

Now, thanks to Omnibus, there's an easy way to distribute your Ruby gems as standalone, full-stack programs. This means folks without Ruby can have as smooth of an experience with your hip new gem as a hardened Rubyist.
Back in 2010, we recommended using bash -l -c to run Ruby executables in Cron that needed the RVM environment. This was a brute approach, and there's a better way: shell scripts. Here's how we suggest running a Ruby executable with RVM and/or Bundler in Cron today.
We spent all of last week clearing out the deadweight that's accumulated in Scout: removing unused CSS rules, database tables + columns, views, and assets plus making our tests run faster. See how we approached it.
Scout's realtime charts were fun to implement and have been rock-solid so far. We were able to implement all of it in Ruby (outside the Javascript used to handle websockets in the browser) and didn’t need to deploy any infrastructure. Learn how we built realtime charts with (mostly) Ruby.
We were writing Objective-C when we started building Redwood. When we released the app this week, almost all of our code was written in Ruby. There were three spots where using MacRuby really made a difference.
We just released the scout_api gem to query metrics stored on Scout. An example: Scout::Server.first(:name => "Web Server").metrics.maximum(:name => 'Memory Used')
Any MacRuby Apps in the App Store? Yes.
While MacRuby isn't available as a public framework on Snow Leopard, there's nothing that prevents you from embedding the framework in your application and publishing it in the Mac App Store. Three examples of MacRuby apps available on the App Store.
Traffic to a Rails app can pass through a number of queues (load balancer, Apache, Passenger, etc). It's easy to miss a backlog in one of these queues - learn how to find and fix queue backlogs in your web stack.
A baby is a lot like a Rails application: the problem is caring for it, not creating it. Where are the caretakers for this Rails business baby boom?
When it comes to monitoring a cluster of servers, there are lots of options with overlapping features. Here's how we're happily monitoring our Rails cluster at Scout.
The author of MongoMapper says Mongo is pretty much his default database. The post also introduces John's Scout plugin for monitoring MongoDB.
Request Log Analyzer, which analyzes a Rails log file and reports throughput and request times, adds Oink file format processing for memory usage metrics. Also includes information on memory leak alerts using Scout.
Jesse Newland of Rails Machine shares three reasons they've started using Resque for high-volume background job processing.
Performance looking good on the server but not in real-life? The warning signs of an under-provisioned Rails application (and how to track it).
Lukas Rieder, Alexander Lang and Eric Lindvall have created a Scout plugin for monitoring Delayed::Job. With the plugin, you can graph and get alerts on running jobs, scheduled jobs, failed jobs and more.
Jesse Newland, CTO of RailsMachine, shares some of his expertise on Phusion Passenger tuning on the Scout Blog.
Request-Log-Analyzer, a command line tool that analyzes logfiles and generates performance reports has lots of great stuff in the 1.4.0 release. Support for more log formats, standard deviation reporting, speed improvements, and more.
This post shows how we (Scout) tackled memory leaks and dying processes in our Scout Agent.
3 pillars of Rails monitoring
The 3 ways we breakdown Rails monitoring at Scout and the tools we use to do it.
Using the new Passenger Monitor plugin, Scout keeps track of key Passenger statistics and alerts you if any of these stats cross a line you indicate.
4 simple steps to detect & fix slow Rails requests - a tutorial on how we use Scout & the Query Reviewer Plugin to quickly fix slow web requests.