A side effect of elben's change to force a wait on every stop call: now that we wait for the process to die, all of our block_* functions are redundant where stop is called since the rtp objects won't return until they actually stop. Wasn't the point to allow asynchronous stopping, but throw an error if they took too long?
We need to come up with a better way to deal with waiting for the processes to die. Is blocking forever the correct way? Should we increase the timeout duration in the cases where we stop?
Currently, we've moved over to blocking 'stop's. They will block until the process dies, possible never releasing.
A side effect of elben's change to force a wait on every stop call: now that we wait for the process to die, all of our block_* functions are redundant where stop is called since the rtp objects won't return until they actually stop. Wasn't the point to allow asynchronous stopping, but throw an error if they took too long?
We need to come up with a better way to deal with waiting for the processes to die. Is blocking forever the correct way? Should we increase the timeout duration in the cases where we stop?
Currently, we've moved over to blocking 'stop's. They will block until the process dies, possible never releasing.