weblogic中的内存泄漏问题

Question

我们有内存泄漏问题。当使用Eclipse Memory分析器分析堆转储（使用应用程序6个小时）时，我们具有以下堆栈：

#Problem suspect 1
    170 000 instances of "org.eclipse.persistence.oxm.XMLDescriptor", loaded by "sun.misc.Launcher$AppClassLoader @ 0x6c000d440" occupy 708 287 296 (30,02 %) bytes. These instances are referenced from one instance of "java.util.concurrent.ConcurrentHashMap$Node[]", loaded by "<system class loader>".  
#Keywords of the issue: java.util.concurrent.ConcurrentHashMap$Node[]
    org.eclipse.persistence.oxm.XMLDescriptor
    sun.misc.Launcher$AppClassLoader @ 0x6c000d440

Answer 1

这是您的问题的解决方案：

public void _old_execute（BatchTaskChunk块，int nrRetriesToPerform）引发BatchRecoverableException，BatchUnrecoverableException {NPBatchTaskChunk signalChunk =（NPBatchTaskChunk）块；

    Signal signal = null;
    try {
        signal = getSignalMgr().findById(signalChunk.getSignalId());
        String result = processSignal(signal, processor);
        signal.setResult(result);
        getSignalMgr().archive(signal);
    } catch (NPBatchSignalRejectException e) {
        // log reject msg
        String msg = e.getMessage();
        if (msg.length() > 254) {
            msg = msg.substring(0, 254);
        }
        signal.setResult(msg);
        try {
            getSignalMgr().archive(signal);
        } catch (ServiceException inner) {

            throw new BatchUnrecoverableExceptionImpl(chunk, "Unable to treat signal #" + signalChunk.getSignalId(), inner);//NOPMD
        }
    } catch (Exception e) {
        throw new BatchUnrecoverableExceptionImpl(chunk, "Unable to treat signal #" + signalChunk.getSignalId(), e);
    }
}