[Zope] zserver non responsive

[Dieter Maurer]

Darian V Schramm wrote at 2005-3-10 10:22 -0500:
> ...
>zeo.log:
>
>2005-03-07T05:33:53 ERROR(200) zrpc-conn:127.0.0.1:41661 Error caught in 
>asyncore
>Traceback (most recent call last):
>   File 
>"/home/isarda-s/zope/273/lib/python/ThreadedAsync/LoopCallback.py", line 
>120, in poll
>   File "/home/isarda-s/Python-2.3.4//lib/python2.3/asyncore.py", line 
>390, in handle_read_event
>     self.handle_read()
>   File "/home/isarda-s/zope/273/lib/python/ZEO/zrpc/smac.py", line 145, 
>in handle_read
>   File "/home/isarda-s/Python-2.3.4//lib/python2.3/asyncore.py", line 
>356, in recv
>     raise socket.error, why
>error: (110, 'Connection timed out')

Looks like a severe network problem...

>event.log:
> ...
>Traceback (most recent call last):
>   File "/home/isarda-s/Python-2.3.4//lib/python2.3/asyncore.py", line 
>69, in read
>     obj.handle_read_event()
>   File "/home/isarda-s/Python-2.3.4//lib/python2.3/asyncore.py", line 
>390, in handle_read_event
>     self.handle_read()
>   File "/home/isarda-s/zope/273/lib/python/ZEO/zrpc/smac.py", line 145, 
>in handle_read
>   File "/home/isarda-s/Python-2.3.4//lib/python2.3/asyncore.py", line 
>356, in recv
>     raise socket.error, why
>error: (9, 'Bad file descriptor')

This, too, might indicate a network problem.
Apparently, the socket is no longer in a sane state.

It may also indicate a race condition in the ZEO communication.
Tim has recently (for Zope 2.7.5) fixed several connection
race conditions.

-- 
Dieter