felix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "AnilKumar Attuluri (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FELIX-5665) High CPU usage on org.apache.felix.framework.BundleWiringImpl.findClassOrResourceByDelegation
Date Wed, 26 Jul 2017 17:33:00 GMT

    [ https://issues.apache.org/jira/browse/FELIX-5665?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16101973#comment-16101973

AnilKumar Attuluri commented on FELIX-5665:

I have created a gradle project here: https://github.com/anilutcs/osgi-reflection-test
It has the instructions to build, run (on karaf) and test.

Here are the findings that might be helpful:

Its the apache log4j library that tries to load the class *_sun.reflect.GeneratedMethodAccessor<number>_*
which ends up in the method *BundleWiringImpl.findClassOrResourceByDelegation* which in turn
ends up looking up for that class everywhere (boot delegation, current class loader, search
imports and including the dynamic imports) and then throws a *ClassNotFoundException*.
So obviously its CPU expensive since this happens several times for every app call. Let me
know if there is a hack to delegate this to some other class loader without impacting the
current class loading. Apparently boot delegation didn't work.

> High CPU usage on org.apache.felix.framework.BundleWiringImpl.findClassOrResourceByDelegation
> ---------------------------------------------------------------------------------------------
>                 Key: FELIX-5665
>                 URL: https://issues.apache.org/jira/browse/FELIX-5665
>             Project: Felix
>          Issue Type: Bug
>    Affects Versions: framework-5.6.4
>            Reporter: AnilKumar Attuluri
>         Attachments: IMG_1.jpg, IMG_2.jpg
> We have been running some performance tests to prepare our OSGi bundle (*running in Apache
Karaf*) for production.
> Just to give some background about our OSGi bundle, we converted an existing Spring application
into an OSGi bundle with all the current dependencies packaged into the bundle as an uber
> When we run >= 500 TPS (each of these calls results in a http call made via a library)
we run into this high CPU usage spikes reaching up to 100% CPU. Please see the image attached,
the spikes in the image are 100% CPU usage while the average is about 40%. Also see the CPU
sampler image which points to *org.apache.felix.framework.BundleWiringImpl.findClassOrResourceByDelegation*
> Is there an existing bug/documentation that already captures this?
> We don't see this behavior when we run the same app in standalone JVM.

This message was sent by Atlassian JIRA

View raw message