Hi all

I have sent an email to zodb-dev about this but I thought I'd see if people
here could shed some light on the issue as well. This time I will state the
issue form a more Zope-client perspective.

Basically after an arbitrary amount of time I start getting these errors in
the Zope client event logs:

2012-10-24T15:33:38 INFO ZEO.ClientStorage storage_c14_zeostorage Testing
connection <ManagedClientConnection ('', 8101)>
2012-10-24T15:33:48 ERROR ZEO.zrpc (1021) CW: error in testConnection
(('', 8101))
Traceback (most recent call last):
line 595, in test_connection
    self.preferred = self.client.testConnection(self.conn)
line 565, in testConnection
    stub = self.StorageServerStubClass(conn)
line 386, in stub
    raise ValueError("Timeout waiting for protocol handshake")
ValueError: Timeout waiting for protocol handshake

I understand that it is possibly Zeo not providing a timely response but
I'm just wondering under what circumstances does the Zope client need to
keep testing the connection? It was my understanding that it connects to
Zeo and then it's connected? Even when the client reports these errors it
is still responding to web requests fine.

It has the knock-on effect that the zeoserver starts consuming more file
descriptors every time these connection attempts occur. There is a
theoretical maximum number of file descriptors that can be used so if the
trend continues then eventually Zeo will start failing to respond to
queries. This is not ideal as the system should be able to run unattended
indefinitely (normal housekeeping operations aside).

Any ideas?


Tim Godfrey
Obsidian Consulting Group

P: +61 3 9355 7844
F: +61 3 9350 4097
E: t...@obsidian.com.au
W: http://www.obsidian.com.au/
Zope-Dev maillist  -  Zope-Dev@zope.org
**  No cross posts or HTML encoding!  **
(Related lists -
 https://mail.zope.org/mailman/listinfo/zope )

Reply via email to