Hung JVM consuming 100% CPU

Jon Masamitsu jon.masamitsu at oracle.com
Tue Mar 1 14:44:39 PST 2011



On 03/01/11 14:00, Charles K Pepperdine wrote:
>
> ...
>> The CMS concurrent collections are fast from what I've seen (on the 
>> order of
>> the time between ParNew collections).   The rate at which objects are 
>> getting
>> promoted is also low (maybe 3m per ParNew collection).  CMS thinks it
>> can wait to start a concurrent collection.  The fact that a 
>> "promotion failure"
>> happened makes it look like fragmentation.
>
> Even with occupancy greater than 90%? I could very well be wrong about 
> this but I thought that promotion failure meant just that, the young 
> gen collector wasn't able to stuff objects up into tenured space. That 
> didn't preclude fragmentation but in this case occupancy is 90% which 
> IIRC is well above the OMG, I can't keep up lets trigger a parallel 
> collection threshold.
>
>
I don't see CMSInitiatingOccupancyFraction on the command line.


        -Xmx896m -Xss128k -XX:NewSize=384M -XX:MaxPermSize=96m
        -XX:+UseParNewGC -XX:+UseConcMarkSweepGC

so CMS isn't using occupancy of the tenured gen to start a cycle. That 
is the
default behavior.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/hotspot-gc-dev/attachments/20110301/525fcdd5/attachment.html 
-------------- next part --------------
_______________________________________________
hotspot-gc-use mailing list
hotspot-gc-use at openjdk.java.net
http://mail.openjdk.java.net/mailman/listinfo/hotspot-gc-use


More information about the hotspot-gc-dev mailing list