This error comes from Windows Socket API. It looks like Windows is having problems with TCP connections, consequently the backlog becomes too big for the buffers to handle. I guess this could be caused by the size of the transaction too, not just by the number of connections, I'm not sure.
Which OS exactly, which version of postgresql? You're connecting to postgresql locally, on the same Windows machine? How much connections does your database get? Are you seeing any TCP/IP errors in your event log? There should be some... Regards, Ales On Thursday, September 20, 2012 2:29:37 AM UTC+2, weheh wrote: > > I'm getting a web2py crash. Here's the trace at the web2py 1.99.7 > (2012-03-04 22:12:08) stable console: > > DEBUG: connect attempt 1, connection error: > Traceback (most recent call last): > File "C:\web2py\gluon\dal.py", line 5955, in __init__ > self.adapter = ADAPTERS[self._dbname](*args) > File "C:\web2py\gluon\dal.py", line 2010, in __init__ > self.pool_connection(connect) > File "C:\web2py\gluon\dal.py", line 485, in pool_connection > self.connection = f() > File "C:\web2py\gluon\dal.py", line 2009, in connect > return self.driver.connect(msg,**driver_args) > OperationalError: could not connect to server: No buffer space available > (0x00002747/10055) > Is the server running on host "localhost" and accepting > TCP/IP connections on port 5432? > cold not connect to server: No buffer space available (0x00002747/1055) > Is the server running on host "localhost" and accepting > TCP/IP connections on port 5432? > > This is the third time in 72 hours it's done this. It's taking down my > site. Anybody have any ideas? > --
