Pretty much what Yoav said ;). If the servlet throws an exception out of the service method, Tomcat can't be sure that the input stream was fully read, so it doesn't know what the next data it's going to read is. In the case of an Exception, it is (usually) a clean shutdown (Tomcat sends a 'Connection: close', and returns the Thread to the pool).
"Shapira, Yoav" <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED] Howdy, A typical pool implementation attempts to guarantee that poolable objects handed out to clients are valid. If an exception occurs in a pool object it's a legitimate strategy to discard that object and create a new valid one, so that a corrupt object will never be handed out to applications. Yoav Shapira Millennium ChemInformatics >-----Original Message----- >From: Rau NF [mailto:[EMAIL PROTECTED] >Sent: Tuesday, August 26, 2003 3:58 PM >To: [EMAIL PROTECTED] >Subject: Re: tomcat - request thread pool question > >Hi - I was referring to the HTTP 1.1 Connector. Thanks >for the information but I am a bit puzzled as to why >the threads should go away (ie die) on getting an >exception. On a heavily loaded server, we can have >different IO exceptions being generated all the time >and the pool will be too busy trying creating new >threads. Not really pool like behaviour ...thoughts >anyone ? > >Thanks > > >==================================================== >It really depends on if you are talking about the HTTP >Connector, or >the AJP >Connector. > >For the HTTP Connector, threads usually have short >lives (i.e. they >exit >after the last Keep-Alive has been handled). >Exceptions thrown out of >the >servlet, and certain HTTP status codes will also end >the thread (since >the >socket stream is likely in an undefined state in these >cases). > >For the AJP Connector, threads usually have long >lives. If you haven't >configured a connectionTimeout, then they usually last >as long as the >Apache >child that they are talking to does. However, if they >get a >SocketException >on output, then it will exit early (since Apache isn't >talking to it >anymore). >============================================= > >--- Rau NF <[EMAIL PROTECTED]> wrote: >> Hi - Under what conditions can the tomcat thread >> pool >> implementation decide to let a thread exit ? >> >> Would this happen if some application code threw an >> exception and is caught only by the tomcat framework >> ? >> Can it happen if the servlet is writing data out to >> a >> connection and gets a SocketException or one of >> those >> IllegalStateExceptions ? eg., >> >> 2003-08-22 01:13:11 ApplicationDispatcher[] >> Servlet.service() for servlet jsp_servlet.some_jsp >> threw exception >> java.lang.IllegalStateException: Cannot create a >> session after the response has been committed >> at >> >org.apache.coyote.tomcat4.CoyoteRequest.doGetSession(CoyoteRequest.java :188 >4) >> at >> >org.apache.coyote.tomcat4.CoyoteRequest.getSession(CoyoteRequest.java:1 731) >> >> > >__________________________________ >Do you Yahoo!? >Yahoo! SiteBuilder - Free, easy-to-use web site design software >http://sitebuilder.yahoo.com > >--------------------------------------------------------------------- >To unsubscribe, e-mail: [EMAIL PROTECTED] >For additional commands, e-mail: [EMAIL PROTECTED] This e-mail, including any attachments, is a confidential business communication, and may contain information that is confidential, proprietary and/or privileged. This e-mail is intended only for the individual(s) to whom it is addressed, and may not be saved, copied, printed, disclosed or used by anyone else. If you are not the(an) intended recipient, please immediately delete this e-mail from your computer system and notify the sender. Thank you. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
