We received some great community feedback on the first TorqueBox 3 beta release and decided to spin a second beta to make sure all the kinks are worked out. If everything goes well we hope to jump from 3.0.0.beta2 straight to 3.0.0 before the end of August. If you haven't yet tried upgrading your application from TorqueBox 2.3.2 to TorqueBox 3 please do and let us know of any issues.
- Download TorqueBox 3.0.0.beta2 (ZIP)
- Download TorqueBox 3.0.0.beta2 (JBoss EAP overlay)
- Browse Getting Started Guide
- Browse HTML manual
- Browse JavaDocs
- Browse Gem RDocs
- Download PDF manual
- Download ePub manual
What is TorqueBox?
TorqueBox is a Ruby application server built on JBoss AS7 and JRuby. It supports Rack-based web frameworks and provides simple Ruby interfaces to standard enterprisey services, including scheduled jobs, caching, messaging, and services.
Highlights of changes in 3.0.0.beta2
Zero downtime deployment improvements
Zero downtime deployments now wait until the new JRuby runtimes are completely initialized before sending requests to them, preventing any web timeout issues you may have seen in 3.0.0.beta1.
If the new runtimes fail to initialize we also log an error and continue to use the old runtimes so that an application isn't left in an unusable state.
Our shipped Capistrano recipes were updated to point TorqueBox at the "current" symlink instead of an actual release, allowing you to use Capistrano to push a new release and then have Capistrano touch tmp/restart.txt to execute a zero downtime deployment. If you've modified our default recipe read the comments in TORQUE-1121 to see the commit details.
OpenShift quickstart updated for TorqueBox 3
We've spent a lot of time testing and tweaking the TorqueBox quickstart for OpenShift. In addition to getting it working with TorqueBox 3, we're also putting the finishing touches on clustering support and will write up how to do that in the next week. If you'd like to run TorqueBox on a PaaS this is the best option.
Heroku instructions updated for TorqueBox 3
We also updated the instructions for running TorqueBox on Heroku. Things are a bit better than with TorqueBox 2, but Heroku's 60 second deploy timeout still poses a challenge for most TorqueBox applications.
torquebox-console updated to work with TorqueBox 3
We released torquebox-console 0.3.0 a week ago that fixes compatibility with TorqueBox 3 while also retaining compatibility with TorqueBox 2. If you previously tried to use torquebox-console with TorqueBox 3.0.0.beta1 or later updating to torquebox-console 0.3.0 should fix things.
Upgrading from 3.0.0.beta1
No changes (other than bumping gem versions) should be required to your application or the AS7 configuration files to upgrade from TorqueBox 3.0.0.beta1 to 3.0.0.beta2.
Don't be a stranger!
As always, if you have any questions about or issues with TorqueBox, please get in touch.
Issues resolved since 3.0.0.beta1
- [TORQUE-1038] - torquebox-console switch_application doesn't work without parentheses.
- [TORQUE-1063] - torquebox run --help doesn't work
- [TORQUE-1071] - torquebox-console doesn't work with 3.x incrementals
- [TORQUE-1112] - Remove mod_cluster recommendation from docs
- [TORQUE-1119] - Remove extra AS7 configs from slim dist
- [TORQUE-1120] - Ensure all documentation config snippets are up-to-date
- [TORQUE-1121] - Figure out how to integrate Capistrano and zero-downtime deploys
- [TORQUE-1122] - torquebox-console broken with 3.0.0.beta1
- [TORQUE-1123] - Disable management auth in the slim dist
- [TORQUE-1124] - Wait until new runtimes are created before handing off zero downtime requests
- [TORQUE-1127] - Update OpenShift quickstart for TB 3
- [TORQUE-1128] - JBoss AS 7.2: Rewrite doesn't support conditions
- [TORQUE-1129] - Abort/Rollback a zero downtime deploy if torquebox cannot start the new runtimes
- [TORQUE-1130] - Race condition with eval'ing of torquebox.rb on slow machines