我不太熟悉J2EE
,因此在解释错误时可能会犯一些错误。请多多包涵。
我试图在我的Java企业应用程序上运行查询,但是glassfish
引发以下异常:
[#|2014-12-05T15:31:00.412+0200|WARNING|glassfishv3.0|javax.enterprise.system.core.transaction.com.sun.jts.CosTransactions|_ThreadID=86;_ThreadName=Thread-1;|JTS5031: Exception [java.lang.RuntimeException: org.postgresql.xa.PGXAException: Error preparing transaction] on Resource [prepare] operation.|#]
[#|2014-12-05T15:31:00.413+0200|SEVERE|glassfishv3.0|javax.enterprise.system.core.transaction.com.sun.jts.CosTransactions|_ThreadID=86;_ThreadName=Thread-1;|JTS5031: Exception [org.omg.CORBA.INTERNAL: vmcid: 0x0 minor code: 0 completed: Maybe] on Resource [rollback] operation.|#]
[#|2014-12-05T15:31:00.439+0200|WARNING|glassfishv3.0|javax.enterprise.system.container.ejb.com.sun.ejb.containers|_ThreadID=86;_ThreadName=Thread-1;|A system exception occurred during an invocation on EJB OFReportTimeoutService method public void com.companyname.appname.service.OFReportTimeoutService.ofTimeout()
javax.ejb.EJBException: Unable to complete container-managed transaction.
at com.sun.ejb.containers.BaseContainer.completeNewTx(BaseContainer.java:4962)
at com.sun.ejb.containers.BaseContainer.postInvokeTx(BaseContainer.java:4716)
at com.sun.ejb.containers.BaseContainer.postInvoke(BaseContainer.java:1941)
at com.sun.ejb.containers.BaseContainer.postInvoke(BaseContainer.java:1892)
at com.sun.ejb.containers.EJBLocalObjectInvocationHandler.invoke(EJBLocalObjectInvocationHandler.java:198)
at com.sun.ejb.containers.EJBLocalObjectInvocationHandlerDelegate.invoke(EJBLocalObjectInvocationHandlerDelegate.java:84)
at com.sun.proxy.$Proxy307.ofTimeout(Unknown Source)
at com.companyname.appname.service.__EJB31_Generated__OFReportTimeoutService__Intf____Bean__.ofTimeout(Unknown Source)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:606)
at com.companyname.appname.servlet.GFv3EJBInvokerJob.execute(GFv3EJBInvokerJob.java:88)
at org.quartz.core.JobRunShell.run(JobRunShell.java:213)
at org.quartz.simpl.SimpleThreadPool$WorkerThread.run(SimpleThreadPool.java:557)
Caused by: javax.transaction.SystemException: org.omg.CORBA.INTERNAL: JTS5031: Exception [org.omg.CORBA.INTERNAL: vmcid: 0x0 minor code: 0 completed: Maybe] on Resource [rollback] operation. vmcid: 0x0 minor code: 0 completed: No
at com.sun.jts.jta.TransactionManagerImpl.commit(TransactionManagerImpl.java:330)
at com.sun.enterprise.transaction.jts.JavaEETransactionManagerJTSDelegate.commitDistributedTransaction(JavaEETransactionManagerJTSDelegate.java:169)
at com.sun.enterprise.transaction.JavaEETransactionManagerSimplified.commit(JavaEETransactionManagerSimplified.java:843)
at com.sun.ejb.containers.BaseContainer.completeNewTx(BaseContainer.java:4951)
... 14 more
|#]
具有
glassfish
,postgresql
和Web应用程序相同版本的另一个系统不会引发任何异常。它们都有相同的domain.xml
和postgresql.conf
文件。我已将
max_prepared_transactions
从100
更改为10000
,并将shared_buffers
从32MB
更改为320MB
,但此操作无效。有任何想法吗?
编辑:
ofTimeout
代码已添加。@Startup
@Singleton(mappedName="OFReportTimeoutService")
public class OFReportTimeoutService {
@EJB
protected QueueManagerService queueManagerService;
@EJB
protected AnalyzerService analyzerService;
@EJB
protected ErrorReportJpaController errorReportJpaController;
@EJB
protected ReportJpaController reportJpaController;
protected Boolean isProcessing = Boolean.FALSE;
private Boolean exceptionPresent = Boolean.FALSE;
private String errorText = "error";
private Integer reportId = 0;
//@Schedule(second="*/10", minute="*", hour="*", persistent=false)
public void ofTimeout() {
//System.out.println("STATE : " + "OfReportTimeOutService is running...") ;
if(exceptionPresent) {
ErrorReport errorReport = errorReportJpaController.create();
errorReport.setErrorDate(new Date());
errorReport.setErrorText(errorText);
errorReport.setReport(reportJpaController.find(reportId));
errorReportJpaController.persist(errorReport);
exceptionPresent = Boolean.FALSE;
if (queueManagerService.getLastReport() != null)
queueManagerService.resetLastReport();
isProcessing = Boolean.FALSE;
}
if (isProcessing)
return;
if (queueManagerService.reportQueueSize() > 0)
isProcessing = Boolean.TRUE;
else {
//System.out.println("STATE : " + "queueManagerService.reportQueueSize == 0 !!!") ;
return;
}
while (queueManagerService.reportQueueSize() > 0)
try {
Report report = queueManagerService.pullReport();
reportId = report.getId();
if ( reportId != null )
System.out.println("STATE : " + "reportId var") ;
analyzerService.process(report);
} catch (ReportJPAException rex) {
Logger.getLogger(OFReportTimeoutService.class.getName()).log(Level.SEVERE, null, rex);
exceptionPresent = Boolean.TRUE;
errorText = rex.toString();
break;
} catch (RuntimeException rex) {
Logger.getLogger(OFReportTimeoutService.class.getName()).log(Level.SEVERE, null, rex);
exceptionPresent = Boolean.TRUE;
errorText = rex.toString();
break;
} catch (Exception ex) {
Logger.getLogger(OFReportTimeoutService.class.getName()).log(Level.SEVERE, null, ex);
exceptionPresent = Boolean.TRUE;
errorText = ex.toString();
break;
}
if(exceptionPresent) {
return;
}
isProcessing = Boolean.FALSE;
queueManagerService.resetLastReport();
}
public Boolean isProcessing() {
return isProcessing;
}
public void setProcessing(Boolean isProcessing) {
this.isProcessing = isProcessing;
}
}
最佳答案
我已经解决了问题。我们正在使用bucardo
复制我们的postgresql数据库,这是导致我出现问题的原因。
在postgresql日志中,我看到了这样的错误日志:
ERROR: cannot PREPARE a transaction that has executed LISTEN, UNLISTEN or NOTIFY
在this blog post中,已经说明了问题的原因:
问题是Postgres LISTEN / NOTIFY系统不能与准备好的事务一起使用。 Bucardo在源表上使用触发器,该触发器发出NOTIFY通知主Bucardo守护程序,其中某些内容已更改,需要复制。但是,他们的应用程序偶尔会发布“准备交易”。因此,他们将更新表,这将触发触发器,并发送NOTIFY。然后,应用程序将发出PREPARE TRANSACTION,该操作会产生上述错误。 Bucardo已准备好处理这种情况;可以将Bucardo守护程序设置为不按指定间隔查找任何更改,而不必使用通知触发器。对于给定的同步,更改Bucardo的行为的步骤很简单:
博客文章上的解决方案不适用于我们。我们负担不起复制导致错误的数据库。因此,我们使用以下命令删除了复制:
[root@Baskan config]# bucardo deactivate synclrms
Deactivating sync synclrms
[root@Baskan config]# bucardo purge synclrms
Purging name synclrms
[root@Baskan config]# bucardo remove sync synclrms
Removed sync "synclrms"
Note: table triggers (if any) are not automatically removed!
由于触发器不会自动删除,因此应手动删除它们:
在我们的案例中,有三个触发器。他们的名字是:
bucardo_delta, bucardo_kick_synclrms, bucardo_note_trunc_synclrms
bucardo_note_trunc_synclrms要删除触发器,请使用以下命令:
drop TRIGGER trigger_name on table_name;
万一bucardo放置的表上可能还有其他触发器,则可以在postgresql中使用以下命令查看表上的所有触发器:
\dS table_name;
完成这些步骤后,系统开始正常运行,没有引发任何异常。