Actually, IE's problem is that it doesn't trust its own web servers
(IIS). The HTTP protocol has request HEAD that will request the headers
associated with a particular request without returning the contents. Older
(I don't remember which) versions of IIS had problems with the HEAD command
not returning the most update to date information (it wouldn't check if an
ASP needed recompiled to check the headers). It seems like the IE group
decided that they would make their browser "better" than Netscape by making
it work for all web servers - broken or not. It will make the request as a
GET, regardless of whether or not it wants the body. Once it has read the
headers it closes the connection (so that it can be smart and save
bandwidth). This closing of the connection before all the content has been
received is what causes the error that you have been seeing.
For small files its possible that the entire file can be in transit
when IE closes the connection and therefore the error message doesn't show
up. This is why this error message doesn't always appear, and appears more
often during times of heavy access (slower response times means fewer
packets in the network). By the way, small, in this case, is a function of
the client's network settings and the server's available resources, but it
should be somewhere less than 2K.
If anyone is interested in a more detailed explanation, feel free to
email off-list.
Randy
> -----Original Message-----
> From: William Kaufman [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, April 19, 2001 12:38 PM
> To: '[EMAIL PROTECTED]'
> Subject: RE: TOMCAT CONFIGURATION
>
>
> > At 09:05 AM 4/19/2001 -0700, you wrote:
> > >Is there any way to stop this occurrence with IE5.X, as a
> > lot of people use this?
>
> Yes: get them all to switch to Netscape 6. (Or, turn the debugging on
> Tomcat all the way off, but I don't recommend that.)
>
> > Isn't this the way all browsers work? I thought
> > that with http once the document length had been
> > passed through the connection it was cut.
>
> Nope: normally, they just load the whole page and close the
> connection,
> which doesn't toss an exception.
>
> What IE seems to be doing is getting the page, seeing (from
> the headers)
> that the version of the file in its cache is new enough, and
> just dropping
> the connection in mid-stream. Normal browsers will simply
> call HTTP GET
> with a last-modified date and let the server figure out if it
> needs to get
> shipped over.
>
>
> -- Bill K.
>
>
> > -----Original Message-----
> > From: Tim O'Neil [mailto:[EMAIL PROTECTED]]
> > Sent: Thursday, April 19, 2001 9:16 AM
> > To: [EMAIL PROTECTED]
> > Subject: RE: TOMCAT CONFIGURATION
> >
> >
> > Isn't this the way all browsers work? I thought
> > that with http once the document length had been
> > passed through the connection it was cut.
> >
> > At 09:05 AM 4/19/2001 -0700, you wrote:
> > >Is there any way to stop this occurrence with IE5.X, as a
> > lot of people use
> > >this?
> > >
> > > > -----Original Message-----
> > > > From: Antonio Vazquez [SMTP:[EMAIL PROTECTED]]
> > > > Sent: Thursday, April 19, 2001 8:53 AM
> > > > To: [EMAIL PROTECTED]
> > > > Subject: RE: TOMCAT CONFIGURATION
> > > > Importance: High
> > > >
> > > > Hi Matias,
> > > > Do you use Internet Explorer 5.X?
> > > >
> > > > In this case, the problem isn´t in the server, the
> problem is that
> > > > Internet
> > > > Explorer cuts the connection with the web server and then
> > Tomcat throws
> > > > that
> > > > exception.
> > > >
> > > > -----Mensaje original-----
> > > > De: Matías Salvador [mailto:[EMAIL PROTECTED]]
> > > > Enviado el: jueves, 19 de abril de 2001 17:45
> > > > Para: [EMAIL PROTECTED]
> > > > Asunto: TOMCAT CONFIGURATION
> > > >
> > > >
> > > > Please, would anybody throw me a hint to config the
> > Tomcat WebServer?
> > > >
> > > > It runs my servlets, but it takes too much to do some
> > requests or some
> > > > responses, besides, it throws me all the time the exception:
> > > >
> > > > 2001-04-17 10:54:33 - ContextManager: SocketException
> > reading request,
> > > > ignored - java.net.SocketException: Connection reset by peer
> > > > at
> java.net.PlainSocketImpl.socketAvailable(Native Method)
> > > > at
> > java.net.PlainSocketImpl.available(PlainSocketImpl.java:451)
> > > > at
> > > > java.net.SocketInputStream.available(SocketInputStream.java:137)
> > > > at
> > > >
> > org.apache.tomcat.service.http.HttpConnectionHandler.processCo
> > nnectionHttp
> > > > Co
> > > > nnectionHandler.java:214)
> > > > at
> > > >
> > org.apache.tomcat.service.TcpWorkerThread.runIt(PoolTcpEndpoin
> > t.java:416)
> > > > at
> > > >
> > org.apache.tomcat.util.ThreadPool$ControlRunnable.run(ThreadPo
> > ol.java:498)
> > > > at java.lang.Thread.run(Thread.java:484)
> > > >
> > > > How can i fix it? How can I optimize it? Is there
> > anything i have to
> > > > change
> > > > in the servlet.xml? What?
> > > >
> > > > Thank you very much!
> > > >
> > > > MATíAS SALVADOR
> > > > JAVA Development Team
> > > > Soluciones Integrales S.A.
> > > > 590 Roca Av. 7° Floor Buenos Aires
> > > > +54.011.4345.0537 int. 132
> >
>
