nlv11281 at natlab.research.philips.com
Mon Apr 10 01:43:37 EDT 2006
> 2. If it is not reasonable, what suggestions do you have for > implementing concurrent access to a resource from (potentially) > multiple interpreters and/or child processes? Run a worker process that manages the resource. Use some form of RPC and IPC to communicate with that process. For many things, you can get way with using a UNIX or TCP socket, sending a request as a pickled Python object (I tend to pickle dictionaries) and getting a response in the same way. The worker process can either run single threaded using a select(), or use some standard mutexing to serialize access to your resource. It would help a lot to know what the mysterious "resource" you want to manage is.