[ZODB-Dev] ZEO client hangs when combined with other asyncore code
stephenmasterman at fastmail.fm
Tue Jun 21 11:54:55 EDT 2005
On Tue, 21 Jun 2005 11:37:34 +0200, zodb at bootsch.demon.nl said:
> Dear people,
> We have an application that makes use of a ZEO client and has other async
> socket code
> Somehow, a call to the ZEO client never returns, it just hangs and sits
Just a stab in the dark, but forget the async code for a minute. Have
you tried just making the simplest possible connection from ZEO client
to the server (across the network, I presume)? For example, at the
interactive Python interpreter create a connection instance and see if
you can get something from the root object.
I once had what might have been a similar problem. ZEO clients on the
same machine as the server worked fine but when the client was on a
different machine I could sort of conenct but never got any response. Or
rather, the client would hang indefinitely when I tried to connect.
(Can't remember what the ZEO server log said.) I seem to recall that it
had something to do with the server configuration and the fact that I
was binding to 127.0.0.1 when I needed to bind to 192.whatever in order
for remote machines to complete a request. Something like that.
Anyway, your problem may have nothing to do with the details of your
application if something like this is happening, so you should verify
that you can successfully connect at all.
> In an attempt to fix this problem we added code to use a different socket
> map for the proxy objects,
> the ZEO client uses the default asyncore socket_map. But that did NOT
> solve our problem.
> At this point we are clueless to the reason behind this behaviour and
> hope that anyone on this
> list has some ideas or similar experiences.
> I realize this is a very terse description of our application, when
> necessary I can elaborate.
> Best regards,
> Paul Boots
> For more information about ZODB, see the ZODB Wiki:
> ZODB-Dev mailing list - ZODB-Dev at zope.org
More information about the ZODB-Dev