[ZODB-Dev] Hanging ZEO-client hangs all other ZEO-clients?

[Paul Winkler]

On Tue, Apr 12, 2005 at 02:10:14PM +0200, Dario Lopez-K?sten wrote:
> Yesterday afternoonI restarted the ZEO server, but today we just now had 
> our first hang again. The logs of the clients state:
> 
> ERROR(200) ZServer uncaptured python exception, closing channel 
> <ZServer.HTTPServer.zhttp_channel connected 127.0.0.1:55055 at 
> 0x1a310774 channel#: 88880 req
> uests:> (socket.error:(104, 'Connection reset by peer') 
> [/usr/local/zope/software/zope-2.6.2-TfDsZeo/ZServer/medusa/asynchat.py|initiate_send|213] 
> [/usr/local/zope/software/zope
> -2.6.2-TfDsZeo/ZServer/medusa/http_server.py|send|417] 
> [/usr/local/zope/software/zope-2.6.2-TfDsZeo/ZServer/medusa/asyncore.py|send|338])
> 
> but it "seems" to fix itself because there are other log entries after 
> that. 

Ignore those.  That just means somebody or something (probably 
a virus or a script kiddie) sent you a malformed request, probably
scanning for known vulnerabilities to exploit.
ZServer just logs the problem and ignores the request.

If you get a lot of those you might consider an http sanitizer
in front of your Zope.  Pound might do.

As for the rest of your troubles... never seen that myself,
afraid I have no idea. 

-- 

Paul Winkler
http://www.slinkp.com