This post originated from an RSS feed registered with Ruby Buzz
by Daniel Berger.
Original Post: The next release of win32-service is nigh
Feed Title: Testing 1,2,3...
Feed URL: http://djberg96.livejournal.com/data/rss
Feed Description: A blog on Ruby and other stuff.
A long standing bug with win32-service may finally be fixed soon. Well, worked around anyway. There was a thread issue that caused a stop event to basically deadlock the service. Oh, it stopped the service, but in a long and ugly way, and the stop event handler wasn't being called.
Patrick Hurley submitted a patch/workaround where we create a separate Ruby thread internally and use that thread to do some polling. Its job is to make sure all other threads that have been spun up have returned when a stop signal is received, then sends a signal to indicate that its finally ready to stop.
It works, though it does require a little extra effort on the end user's part (it requires another state check), which I'm hoping to abstract away in a handy little wrapper.
This is not the final patch. Patrick has an even better one, but I'm going to wait until we separate the Daemon class from the Service class.
Previous Topic
