From owner-freebsd-java@FreeBSD.ORG Sun Aug 17 15:06:29 2003 Return-Path: Delivered-To: freebsd-java@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 657AD37B401 for ; Sun, 17 Aug 2003 15:06:29 -0700 (PDT) Received: from cableaz.com (mail.cableaz.com [66.218.238.20]) by mx1.FreeBSD.org (Postfix) with SMTP id B0A6543F75 for ; Sun, 17 Aug 2003 15:06:28 -0700 (PDT) (envelope-from desertfox@cableaz.com) Received: (qmail 25588 invoked by uid 0); 17 Aug 2003 22:06:27 -0000 Received: from unknown (HELO CopperLeafAngel) (66.218.239.229) by 0 with SMTP; 17 Aug 2003 22:06:27 -0000 From: Erick Smith To: freebsd-java@freebsd.org Date: Sun, 17 Aug 2003 15:06:24 -0700 User-Agent: KMail/1.5.3 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200308171506.24910.desertfox@cableaz.com> Subject: Large ClientServer app...some info X-BeenThere: freebsd-java@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Porting Java to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 17 Aug 2003 22:06:29 -0000 OK, at this time I can't be too specific, but I'm working on it. The reason I can't be too specific is because the jdk doesn't actually report any errors, it simply hangs. The Details: This is a client-Server app with the client delivered via an Applet (JApplet actually) in a web page. Accessible from the internet. You can check it out yourself at: http://www.manifestdestinygame.com/testMD-BSD.html Bear in mind the version running on the web page is using the linux-sun-jdk1.4.1 because this works, albeit slow. When using the native jdk, the server locks up before clients can even access it. It locks up parsing text files using the tokenizer. It hangs usually during the second use of my parser class (I can make the code accessible if you need it, it's not too complicated). However, if I do less reporting to the console, the parser will last til the 3rd run through (3 seperate txt files, same parser code), so perhaps this is a garbage collection issue. The text files are quite long (>10000 entries in some cases), and the Parser seems to lock generally after about 3000 tokens processed, again though, this can be manipulated to hang during the second or third pass. The client side hangs waiting for information from the socket. I'm not sure how relavant this is though, because the only time I can get the native-jdk to connect is when the server is using the linux-sun base because otherwise the server hangs before sockets are set up. Again, the JVM says nothing to the console or otherwise, the system just hangs. I hope this helps some. As I said, I'd be willing to post code if necc., so long as it isn't redistributed for any other purpose than fixing the native jdk. Thanks, Erick Smith