Gabriel Gruber pisze: > > Hello, Hi.
> I am facing a strange issue with our product, which was recently ported > to cocoon 2.2 and is now in the final stabilization phase. It seems that > for some reason various exceptions occure on the log which might have > something to do with servlet service framework or inproper configuration > of jetty. this particular exception is repoduceable as we get it, when > driving a seleninium testcase against an embedded jetty/hsqldb > configuration of our product. > > Interestingly this does not have any bad consequence in the frontend. > however something seems to go wrong... <snip/> > Caused by: java.net.SocketException: Connection reset by peer: socket > write error > > at java.net.SocketOutputStream.socketWrite0(Native Method) > > at > java.net.SocketOutputStream.socketWrite(SocketOutputStream.java:92) > > at > java.net.SocketOutputStream.write(SocketOutputStream.java:136) > > at > org.mortbay.io.ByteArrayBuffer.writeTo(ByteArrayBuffer.java:169) > > at > org.mortbay.io.bio.StreamEndPoint.flush(StreamEndPoint.java:122) > > at > org.mortbay.jetty.HttpGenerator.flush(HttpGenerator.java:693) > > ... 82 more > > Does anyone have a clue, what could be the issue? I believe the rest of the stack trace is not relevant thus I wouldn't blame SSF or Cocoon in general. The message "Connection reset by peer: socket write error" suggests more a problem with Jetty or application accessing jetty. Basically it looks like someone is closing connection too early so the rest of data cannot be written. Yet, it's interesting that it does not affect the client and the data it receives. I would suggest testing this with different clinet/servlet container configuration. Also, does it happen with responses having 200 (HTTP_OK) status code? -- Grzegorz Kossakowski --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
