Thanks for the detailed repsone... sorry for the lag in responding to it. After reading and further thought, the only reason I was using setdefaulttimeout in the first place (rather then using a direct settimeout on the socket) was because it seemed like the only way (and easy) of getting access to the seemingly deeply buried socket being used by xmlrpclib. That was prior to me using threads of course. I then started trying to make this solution work with thread, but it is now too convoluted as you say. Now I think the best solution is likely to redirect my efforts at getting access to the socket used by xmlrpclib so that I can set it's timeout directly. I'm still unclear how to do this cleanly, though.
Getting to some of your comments. > When you say "one thread affects another", I see that your example uses > the same function for both threads. IMHO it's much better to override > the thread's run() method than to provide a callable at thread creating > time. That way you can be sure each thread's execution is firmly in the > context of the particular thread instance's namespace. > > having said all this, I don't think that's your issue. Correct - the bottom code is nothing to do with my code and was only to quickly prove that it was cross-thread. > This seems extremely contorted, and I'm pretty sure we can find a better > way. Couldn't agree more! > The threads' network calls should be yielding process control during > their timeout period to allow other runnable threads to proceed. That's Yep. This is not causing me any problem. > You are aware, I presume, that you can set a timeout on each socket > individually using its settimeout() method? Yes, but I momentarily had forgot about it... as mentioned I ended up making the since-bad choice of using setdefaulttimeout to get timeouts set on the inaccessible sockets. Then I carried it too far... > See above. However, this *does* require you to have access to the > sockets, which is tricky if they are buried deep in some opaque object's > methods. Any help on how to crack the safe would be appreciated. > There are locks! I suspect what you need is a threading.Rlock object, > that a thread has to hold to be able to modify the (global) default > timeout. This isn't a full solution to your problem, though, as you have > correctly deduced. Not quite what I was after I don't think since potentially interfering code needs to check the lock (via acquire) to avoid conflict. What I guess I mean is something general for the process saying "never ever interrupt this block og code by running code on another thread, regardless of whether the other thread(s) check a lock". Thinking more about it it seems unreasonable so I'll drop the question. Russ -- http://mail.python.org/mailman/listinfo/python-list
