The week ending 13 February 2011 was a good week. The snows came to an end, the weather turned at least teasingly warmer, and much work was performed on TorqueBox.
MagicRuby, yet again
Since the last update, Ben Browning did a fine job of post-production on the film he and Toby Crawley shot at MagicRuby. Jim posted it, along with some personal notes. If you were unable to attend MagicRuby, you can watch Jim's complete presentation online, now, for free.
News & Articles
As promised, Ben blogged about highly-available services with failover, showing just how very simple it is to deploy an HA service on a TorqueBox cluster.
Toby also wrote about advanced options now available with messaging, such as message priority and TTL.
This week saw the three great minds of Jim, Ben and Toby working on various aspects of messaging. Jim extended Toby's work on backgroundable methods to support non-Rails deployments, while Toby worked on supporting various edge-cases related to development-mode reloading of classes.
Ben has been working on making our synchronous queues more performant and conscious of resource usage. Additionally, he eliminated some peformance issues in our web session-hanlding implementation.
I (Bob) enabled deployment of Ruby apps lacking a web component. I'll be blogging about a-la-carte knobs this week. At the logical insistence of Jim, I also began work on improving our admittedly crappy rubygem names. Plus, virtual host support has been fixed in the latest development build.
The team all uses last.fm a fair bit, and if you'd like to listen to the music that's ingrained in this code, check out our team station on last.fm. The whole team is hoping to get together at the Red Hat Summit/JBossWorld conference this May. Come find us if you're in the area.