Jesse Noller <[EMAIL PROTECTED]> added the comment: Ben's incref error != the connection refused error. Here's some information from Scott Leerssen who is using the original .52 code:
Hi Jesse, I am having some trouble with the processing module (0.52) and thought you might be of some help since you were so involved with it in the PEP process. I have a queue of Process objects from which a runner thread pulls and starts processes. The producers for the queue are many, many threads. I'm seeing some very odd behavior when lots of processes try to start while others are being reaped at the same time. I see tracebacks that look like the one below, and also occasionally get "hangs" where the process module "seems" to take over the GIL and never give it back. I did find a race condition bug in process.py where a _cleaner function is run in start(), potentially joining on processes that other threads may also be joining at the same time, which leads to KeyError exceptions coming from _cleaner. I mutexed all my starts and joins to get around that for now, but am stumped on the issue below. Does it look familiar to you? Thanks, and feel free to shrug and reply "beats me". Scott [dmbd:32451:-1236096096] 2008-07-31 16:52:10,609 SYSMESG Process Process-24: [dmbd:32451:-1236096096] 2008-07-31 16:52:10,612 SYSMESG [dmbd:32451:-1236096096] 2008-07-31 16:52:10,613 SYSMESG Traceback (most recent call last): [dmbd:32451:-1236096096] 2008-07-31 16:52:10,613 SYSMESG File "/opt/race/share/sw/os/Linux_2.6_i686/python/lib/python2.5/site- packages/processing/process.py", line 224, in _bootstrap [dmbd:32451:-1236096096] 2008-07-31 16:52:10,615 SYSMESG _runAfterForkers() [dmbd:32451:-1236096096] 2008-07-31 16:52:10,616 SYSMESG File "/opt/race/share/sw/os/Linux_2.6_i686/python/lib/python2.5/site- packages/processing/process.py", line 300, in _runAfterForkers [dmbd:32451:-1236096096] 2008-07-31 16:52:10,617 SYSMESG func(obj) [dmbd:32451:-1236096096] 2008-07-31 16:52:10,617 SYSMESG File "/opt/race/share/sw/os/Linux_2.6_i686/python/lib/python2.5/site- packages/processing/managers.py", line 806, in _afterFork [dmbd:32451:-1236096096] 2008-07-31 16:52:10,620 SYSMESG self._incref() [dmbd:32451:-1236096096] 2008-07-31 16:52:10,621 SYSMESG File "/opt/race/share/sw/os/Linux_2.6_i686/python/lib/python2.5/site- packages/processing/managers.py", line 760, in _incref [dmbd:32451:-1236096096] 2008-07-31 16:52:10,621 SYSMESG dispatch(connection, None, 'incref', (self._id,)) [dmbd:32451:-1236096096] 2008-07-31 16:52:10,622 SYSMESG File "/opt/race/share/sw/os/Linux_2.6_i686/python/lib/python2.5/site- packages/processing/managers.py", line 92, in dispatch [dmbd:32451:-1236096096] 2008-07-31 16:52:10,623 SYSMESG raise result [dmbd:32451:-1236096096] 2008-07-31 16:52:10,624 SYSMESG RemoteError: --------------------------------------------------------------------------- Remote Traceback (most recent call last): File "/opt/race/share/sw/os/Linux_2.6_i686/python/lib/python2.5/site- packages/processing/managers.py", line 190, in handleRequest result = func(c, *args, **kwds) File "/opt/race/share/sw/os/Linux_2.6_i686/python/lib/python2.5/site- packages/processing/managers.py", line 404, in incref self.id_to_refcount[ident] += 1 KeyError: 2880341100L _______________________________________ Python tracker <[EMAIL PROTECTED]> <http://bugs.python.org/issue3419> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com