On 27 January 2012 14:43, Mark Collin <[email protected]> wrote: > The tests all seem to be completing correctly and showing pass/fail as > appropriate. > > I have checked the thread count and the threads are being closed off after > this message appears so I presume it's just not shutting the threads down in > time (I am running this on a sub optimal machine so I don't know if you > would see that same messages with something a bit more powerful). > > For now I'll assume that it is expected behaviour and worry about it more if > the tests start behaving in an unexpected way. >
I've added a property to control the timeout: URL: http://svn.apache.org/viewvc?rev=1236708&view=rev This will be in the upcoming JMeter release. > > -----Original Message----- > From: sebb [mailto:[email protected]] > Sent: 27 January 2012 14:05 > To: [email protected] > Subject: Re: JVM did not exit; following threads are still running > (DestroyJavaVM is OK): > > On 27 January 2012 12:20, Mark Collin <[email protected]> wrote: >> I'm occasionally seeing the following printed out to console when >> running JMeter through maven: >> >> >> >> JVM did not exit; following threads are still running (DestroyJavaVM is > OK): > > Are there any other messages following this? > >> >> >> Looking through the codebase it seems to be part of a general clean up >> after a test, however is this something I should expect to see now and >> again or is it an indication of something more sinister. I can >> supress it easy enough but I don't want to do that if it's an >> indication of something serious going wrong. > > It just indicates that JMeter could not exit normally because there are > still some non-daemon threads running. > > Previously, JMeter used to call System.exit(), but that causes problems when > running from another application (e.g. Maven), so the code was changed to > just end the JMeter threads. > > The message was added to provide feedback to the user in case the JVM did > not exit as expected. > This is mainly useful when using client-server or non-GUI mode. > It is implemented as a daemon thread which waits a short while; if the > thread is able to continue, clearly the JVM did not exit, so it prints the > message. > > But where JMeter is being invoked by another application which continues to > run after the test finishes, the message is not all that useful. > It should perhaps be made optional; also perhaps the timeout should be > configurable (currently 2000ms). > > In your case, perhaps Maven is sometimes taking longer to complete than the > timeout (2s) in the JMeter daemon thread that prints the message. > >> >> >> >> -- >> This message contains confidential information and is intended only for > the individual named. If you are not the named addressee you should not > disseminate, distribute or copy this e-mail. Please notify the sender > immediately by e-mail if you have received this e-mail by mistake and delete > this e-mail from your system. If you are not the intended recipient you are > notified that disclosing, copying, distributing or taking any action in > reliance on the contents of this information is strictly prohibited. >> >> If you have received this email in error please notify >> [email protected] > > -- > This message contains confidential information and is intended only for the > individual named. If you are not the named addressee you should not > disseminate, distribute or copy this e-mail. Please notify the sender > immediately by e-mail if you have received this e-mail by mistake and delete > this e-mail from your system. If you are not the intended recipient you are > notified that disclosing, copying, distributing or taking any action in > reliance on the contents of this information is strictly prohibited. > > If you have received this email in error please notify > [email protected]
