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.
Created attachment 119635 [details] Logs and snapshot all in zip file The IDE stops responding due to 99% CPU use. There are Lucene exceptions in the log. The whole computer becomes unresponsive and it takes 30 minutes to create and save a snapshot. Impossible to continue using this build for any work. When deploying a web application to GlassFish, there is a tab in the output window for this "<Project name> (run-deploy)" but the content is empty and the process does not show any output in the output window for a long time. CPU is at 99% and in Windows XP it takes minutes to start the task manager. NetBeans is still responding at a glacial speed. It is NetBeans not GlassFish consuming the CPU. I click the button "Profile the IDE" and after a minute I click it again. Saving snapshot does not want to finish, either. The deploy task is still running after 30 minutes or so and has produced output: ant -f <my project> -DforceRedeploy=true -Ddirectory.deployment.supported=true run-deploy init: deps-module-jar: deps-ear-jar: deps-jar: The deploy task may not be what is causing this because saving snapshot does not finish, either. I don't want to lose the snapshot while memory use is going up so I cancel the running deploy task. ant -f <my project> -DforceRedeploy=true -Ddirectory.deployment.supported=true run-deploy init: deps-module-jar: deps-ear-jar: deps-jar: BUILD STOPPED (total time: 41 minutes 44 seconds)
netbeans-trunk-nightly-201205160400-ml-windows.exe
The IDE was locked due to high IO load and basically no operations on disk were performed during that time. I suspect Glassfish deployment but it can be any process blocking the IO on Windows machine. Re-assigning to glassfish for evaluation. The Lucene exceptions are unrelated to the lockup.
Is it happening any time and with any kind of deployment? What GF (version, local/remote) do you use? We know that there is JAX-WS bug in GF 3.1.2 which may cause NetBeans to hang when running ant task related to WS. Check 211962 for more details. Knowing more about your project and GlassFish may let me know if you may be facing the same issue or not.
It usually takes less than a minute to deploy the project not over 41 minutes. I does not look so much like a GlassFish issue to me although the GlassFish deployment was the only task that I was aware of at the time. I killed the deploy task from within NetBeans and then the CPU use was still as high as before. If there are no other clues in the log file and in the snapshot then I think there is not much we can do until it happens again. My question would be what other steps I could take while it happens to make your analysis easier. Where was 45 minutes of CPU load going? Into I/O?
Log shows a lot of 'slow execution' warnings and also it looks like they were written also before deployment. In snapshot there are couple of disk I/O that took a very long time. But It was waiting on java.io, not spending time inside NB code. At this point I'm even not sure if this was caused by NetBeans. Are you sure that there was nothing else doing intensive disk I/O running on your windows machine? Next time you can try to check all processes running there. Think http://technet.microsoft.com/en-us/sysinternals/bb896653 (Process Explorer) is a good tool for this.
I'm closing this bug now because I can't do much without knowing more. Please feel free to open it again when you'll the same problem.