…child process when the parent receives a signal
…ker from redis. this causes workers to disappear after the first fork
This commit solved an issue when the child never get killed, and will only reset cant_fork. Therefore, a new fork will be created, over and over. refs: https://github.com/staugaard/resque-multi-job-forks/issues#issue/4
…or timeout limit per fork. Fiddled with unit tests.