This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
Reported by the user: (M9, NB5.0, JDK 1.5.0_04 32bit Client VM, Windows 2003 server 64bit, 4GB RAM, dual Opteron processors, remote profiling) "I have modified our batch script to include the -agent param that is required. The vm complains about not being able to creating the heap. I removed the -Xms and -Xmx params and everything works and I am able to profile from Netbeans. Putting the params back and removing the -agent param causes everything to work as well. The max memory parameter is set to 1442M. I was able to get it to work at 1024M." "This issue is very easy to reproduce: 1. Download the remote profiler dlls and jars. 2. Select any application and set -Xms256M and -Xmx1426M and then add the profiling flag (-agentpath). 3. Start the application and everything should work. 4. Exit the application and change the -Xmx flag to 1427M and try to start the application. The VM should exit and complain about the heap."
Is this reproducible on machines other than Windows 2003 Server?
I've done some experiments, here are the results: Win XP, 1GB RAM, JDK 5.0u4 1. with -client VM - the VM complains if the -Xms/-Xmx is larger than 1398M 2. with -server VM - the VM complains if the -Xms/-Xmx is larger than 1426M In both cases, the same behavior occurs with *and* without the profiler agent, there is no difference in the behavior. The error message is: Error occurred during initialization of VM Could not reserve enough space for object heap Could not create the Java virtual machine.
Tested on Fedora Core 4, 2GB RAM, 2x Opteron 64bit JDK 1.5.0_04 64bit: 1. with -server VM - the VM complains if the -Xms/-Xmx is larger than 5697M JDK 1.5.0_05 32bit: 1. with -client VM - the VM complains if the -Xms/-Xmx is larger than 3658M (3686M with agent) 2. with -server VM - the VM complains if the -Xms/-Xmx is larger than 3684M (3684M with agent)
Will not address in 1.0
Milestone cleanup: future->next
I am not able to reproduce it.