Why does cflock sometimes cause 'timeout value is negative' error?

Question

I have a ColdFusion event gateway that sometimes produces an error on the following line (where Local.curThread is a digit 1-5):

<cflock name="eventCachedQueryUpdate_thread_#Local.curThread#" timeout="0" throwontimeout="no">

The error is:

Message: timeout value is negative
Type: java.lang.IllegalArgumentException

Here's a StackTrace:

java.lang.IllegalArgumentException: timeout value is negative at java.lang.Object.wait(Native Method) at coldfusion.runtime.RWLock.waitForLock(RWLock.java:154) at coldfusion.runtime.RWLock.requestWriteLock(RWLock.java:124) at coldfusion.runtime.RWLock.requestLock(RWLock.java:46) at coldfusion.runtime.LockManager.requestNamedLock(LockManager.java:73) at coldfusion.tagext.lang.LockTag.doStartTag(LockTag.java:186) at cfeventCachedQueryUpdate2ecfc749015300$funcONINCOMINGMESSAGE.runFunction(D:\Applications\CFusion\CustomTags\4C\eventCachedQueryUpdate.cfc:21) at coldfusion.runtime.UDFMethod.invoke(UDFMethod.java:472) at coldfusion.filter.SilentFilter.invoke(SilentFilter.java:47) at coldfusion.runtime.UDFMethod$ArgumentCollectionFilter.invoke(UDFMethod.java:368) at coldfusion.filter.FunctionAccessFilter.invoke(FunctionAccessFilter.java:55) at coldfusion.runtime.UDFMethod.runFilterChain(UDFMethod.java:321) at coldfusion.runtime.UDFMethod.invoke(UDFMethod.java:220) at coldfusion.runtime.TemplateProxy.invoke(TemplateProxy.java:491) at coldfusion.runtime.TemplateProxy.invoke(TemplateProxy.java:437) at coldfusion.filter.EventComponentFilter.invoke(EventComponentFilter.java:67) at coldfusion.filter.ApplicationFilter.invoke(ApplicationFilter.java:399) at coldfusion.filter.EventRequestMonitorFilter.invoke(EventRequestMonitorFilter.java:47) at coldfusion.filter.MonitoringFilter.invoke(MonitoringFilter.java:40) at coldfusion.filter.ClientScopePersistenceFilter.invoke(ClientScopePersistenceFilter.java:28) at coldfusion.filter.GlobalsFilter.invoke(GlobalsFilter.java:38) at coldfusion.filter.DatasourceFilter.invoke(DatasourceFilter.java:22) at coldfusion.eventgateway.EventProxy.invokeComponent(EventProxy.java:66) at coldfusion.eventgateway.EventProxy.invokeComponent(EventProxy.java:47) at coldfusion.eventgateway.EventRequestHandler.invokeCFC(EventRequestHandler.java:185) at coldfusion.eventgateway.EventRequestHandler.processRequest(EventRequestHandler.java:111) at coldfusion.eventgateway.EventRequestDispatcher$Task.run(EventRequestDispatcher.java:122) at coldfusion.util.SimpleWorkerThread.run(SimpleThreadPool.java:210)

Since the timeout attribute of cflock is always 0 I don't see how that could be resulting in a negative timeout value. Any ideas what's causing this and how to prevent it? Is this a bug in cflock?

I'm running Adobe ColdFusion 9.0.1.274733 Enterprise on Windows Server 2008 R2 with Java 1.6.0_17.

Answer 1

If I had to guess, it would be that event gateway requests handle the way timeout="0" incorrectly. The docs for <cflock> for ColdFusion 9 say this:

If you set timout="0", the timeout is determined by the "Timeout Requests after x" setting in the ColdFusion Administrator Settings page, if that setting is enabled. However, if the setting is not enabled, and you set timeout="0", ColdFusion can wait indefinitely to obtain the lock.

Perhaps - and this really is just speculation - CF deals with locks that time out incorrectly when run via an event gateway?

To troubleshoot, maybe put a sleep() in the locked code so that concurrent gateway requests will timeout, and see what gets logged?

Do you have a request timeout set?