Red Hat Bugzilla – Bug 592553
java applets max out the CPU
Last modified: 2010-08-05 07:31:43 EDT
Description of problem:
Java applets seem to max out the CPU now - two examples are www.jigzone.com and the file uploading applet at adrive.com. There were updates to java-1.6.0-openjdk and java-1.6.0-openjdk-plugin two days ago so these seem to be likely culprits.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. Go to www.jigzone.com.
Java applet maxes out CPU continuously, process must be killed to stop it.
Reasonable CPU usage.
Sorry, more specifically you should go to a page such as
which loads a java applet.
Well, it turns out that the version of java-1.6.0-openjdk is NEWER in F11 (-35.b18.fc11) and F12 (-38.b18.fc12) than in F13 or Rawhide (both -37.b17.fc13), and the problem only exists in the newer F11 and F12 versions. I see it also in F12 i386, and I think also in F11 i386 (not sure since my F11 box is too slow to allow the jigzone applet to load in time, but I see the high java CPU usage there as well).
This seems to be a problem introduced into b18
I also suffered 100% CPU usage and applet failed to initialise after updating to java-1.6.0-openjdk-184.108.40.206-35.b18.fc11
Firefox locks up and the proccess has to be killed.
The updated openjdk-plugin also fails in Konqueror andr rekonq.
This new problem manifested on the java applet here:
To test - you need to first register an email address with the website, but it is free and almost immediate.
Normal behaviour was restored when I reverted to
I see this also on jigzone.com as suggested by Robatino.
Fedora Bugzappers volunteer triage team
Fedora Bugzappers volunteer triage team
I'm also seeing this with http://www.multicians.org/thvv/gpw.html and java-1.6.0-openjdk-plugin-220.127.116.11-38.b18.fc13.i686 on F-13.
java version "1.6.0_18"
OpenJDK Runtime Environment (IcedTea6 1.8) (fedora-38.b18.fc13-i386)
OpenJDK Client VM (build 14.0-b16, mixed mode)
java.lang.InterruptedException: sleep interrupted
at java.lang.Thread.sleep(Native Method)
*** Bug 592586 has been marked as a duplicate of this bug. ***
Is this problem consistently reproducible?
Also, what is the OS/OpenJDK version combination?
There is currently an issue whereby F12 update has a newer version than the F13 one, causing it to get installed:
100% reproducible for me. Version noted in comment #5. Was a fresh F13 install, not upgrade.
It was 100% reproducible when I was using the newest broken F11 and F12 versions. I don't see it on the older F13 version. I'm aware of the version problem as indicated in comment #2.
There is a nonfunctional (for me, anyway) update submitted for stable at
Please test to make sure that another broken update doesn't get into F13.
See https://bugzilla.redhat.com/show_bug.cgi?id=594550#c4 for info on how to get the broken Java plugin in the now pushed update to work. If nobody sees 100% CPU usage after doing this, I'll close this as CURRENTRELEASE.
FYI the broken links problem has been reported as bug #602906.
I fixed the alternatives link as described above to make the Java plugin work, and the 100% CPU usage is still there. Now that the problem is in F13 as well, bumping the Version to 13. I'll downgrade to the earlier version until a fixed version is released.
These bugs (wrong link and 100% CPU usage) now exist in Rawhide as well.
Should the Version of this bug be bumped to rawhide, or should it be cloned?
java-1.6.0-openjdk-18.104.22.168-41.b18.fc13 has been submitted as an update for Fedora 13.
This update fixes both this bug and bug #602906 for me.
java-1.6.0-openjdk-22.214.171.124-41.b18.fc13 has been pushed to the Fedora 13 stable repository. If problems still persist, please make note of it in this bug report.
This bug was originally reported against Fedora 12. Since F12 is still being maintained, will there be an update for it too?
Reopened and switched Version to 12 until a F12 update is pushed.
It shouldn't be too difficult to build this for F11 and push that, too, before EOL on June 25. I'll switch the component to F11 after the F12 update is pushed - the F12 version is being built right now at
java-1.6.0-openjdk-126.96.36.199-39.b18.fc12 has been submitted as an update for Fedora 12.
java-1.6.0-openjdk-188.8.131.52-36.b18.fc11 has been submitted as an update for Fedora 11.
java-1.6.0-openjdk-184.108.40.206-36.b18.fc11 has been pushed to the Fedora 11 stable repository. If problems still persist, please make note of it in this bug report.
java-1.6.0-openjdk-220.127.116.11-39.b18.fc12 has been pushed to the Fedora 12 stable repository. If problems still persist, please make note of it in this bug report.
The latest java build in F12 is still broken.
I have had to revert to java-1.6.0-openjdk-18.104.22.168-34.b17
to find a build without regression which will run the java applet that I mention in comment 3.
java-1.6.0-openjdk-22.214.171.124-39.b18 locks up firefox.
This is in addition to Bug 594978
Can you reproduce this problem with any applet that doesn't require registration (such as the one I mentioned in comment 1)? The site you mentioned requires more information than I'd like to give (though I could lie, of course), and http://www.formula1.com/termsandconditions.html is a bit long. Additionally, if it doesn't involve 100% CPU, then it's not really this bug, so you might want to file a new one. You might also want to recheck bug 594978 since firefox in F12 was recently updated to 3.5.10.
FWIW - I had to wipe out my ~/.java cache before things worked for me.
java-1.6.0-openjdk-126.96.36.199-40.b18.fc12 has been submitted as an update for Fedora 12.