Hi, I'm having the same problem that Ben Tomasini reported a few days ago. Ben, have you been able to solve this?
Running on Ubuntu 7.04 on a dual core machine, JDK 1.5.0_11, maven 2.0.5 and with the MAVEN_OPTS suggested by Jesse. Can't upgrade to maven 2.0.6 because of the Selenium compilation errors reported in another thread. When the tests hang, the jstack tool produces the following thread dumps (some details removed for brevity): Thread 14524: (state = BLOCKED) - java.lang.Object.wait() - org.openqa.selenium.server.SingleEntryAsyncQueue.get() - org.openqa.selenium.server.SeleneseQueue.doCommand() - org.openqa.selenium.server.SeleniumDriverResourceHandler.doCommand() - org.openqa.selenium.server.SeleniumDriverResourceHandler.handleCommandRequest() - org.openqa.selenium.server.SeleniumDriverResourceHandler.handle() - org.mortbay.http.HttpContext.handle() - org.mortbay.http.HttpContext.handle() - org.mortbay.http.HttpServer.service() ... Thread 14501: (state = IN_NATIVE) - java.net.SocketInputStream.socketRead0() - ... - java.io.BufferedInputStream.read() - sun.net.www.http.HttpClient.parseHTTPHeader() - sun.net.www.http.HttpClient.parseHTTP() - sun.net.www.protocol.http.HttpURLConnection.getInputStream() - java.net.HttpURLConnection.getResponseCode() - com.thoughtworks.selenium.HttpCommandProcessor.getCommandResponse() - com.thoughtworks.selenium.HttpCommandProcessor.executeCommandOnServlet() - com.thoughtworks.selenium.HttpCommandProcessor.doCommand() - org.apache.tapestry.test.ErrorReportingCommandProcessor.doCommand() line=68 ... If I repeat the dump a few minutes later, I get different results, the first thread is waiting inside o.o.s.server.SingleEntryAsyncQueue.put() instead of get(), the topmost tapestry method in the second thread is - org.apache.tapestry.test.ErrorReportingCommandProcessor.reportError() line=56 - org.apache.tapestry.test.ErrorReportingCommandProcessor.doCommand() line=72 I haven't looked at any of the source code involved, but from the stack traces I guess that the test is waiting for the server; the server is mostly blocked on some wait() that probably times out occasionally, but not quickly enough for the tests to fail in a reasonable amount of time. Lars PS: Sorry for not replying to the original thread, I just subscribed here --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
