Jonathan Daugherty cygnus at cprogrammer.org
Tue Jan 20 22:48:43 CET 2004

I'm using python 2.3.3 (#2, Jan  4 2004, 12:24:16), [GCC 3.3.3 20031229
(prerelease) (Debian)], debian unstable.  I have written a client and
server program which communicate using the built-in xmlrpclib and
SimpleXMLRPCServer.  The server starts up fine, and the client can
connect to it and issue commands (method calls) just fine.  However,
after some unknown period of time, the client hangs at the time of the
method call, during which time netstat reports:

      recv  send
tcp   307   0   ESTABLISHED

Once I break the client and run netstat again, I get:

tcp   307   0   CLOSE_WAIT

And lines such as these hang around in netstat's output for quite some
time (several are there now, left over from attempts to run the client
about 8 hours ago).  I haven't been able to identify exactly when this
starts to occur or why.  Sometimes it happens soon after starting the
server; sometimes it takes days.

The server is multi-threaded.  The XML-RPC thread is the main thread.
Proper locking mechanisms are used and I have been able to determine
that deadlock is not an issue.

Does anyone have any ideas?



  Jonathan Daugherty

  "It's a book about a Spanish guy called Manual, you should read it."
                                                            -- Dilbert

More information about the Python-list mailing list