tracker issue : CF-4199420

select a category, or use search below
(searches all categories and all time range)
Title:

Large numbers of NullPointerExceptions, every couple of days (session related?)

| View in Tracker

Status/Resolution/Reason: Closed/Withdrawn/CannotReproduce

Reporter/Name(from Bugbase): Terry Ford / Terry Ford ()

Created: 08/03/2017

Components: General Server

Versions: 2016

Failure Type: Incorrectly functioning

Found In Build/Fixed In Build: Current /

Priority/Frequency: Normal / All users will encounter

Locale/System: English / Linux Red Hat 7.1

Vote Count: 2

Problem Description:

The following issue seems to occur every few days.  When it occurs, there are large numbers of exceptions at exactly the same instant.   There were *47* of these NullPointerExceptions in our exception.log this morning:
 
"Error","scheduler-2","08/03/17","05:11:14","","'' The specific sequence of files included or processed is: '''' "
java.lang.NullPointerException
        at coldfusion.runtime.NeoJspWriter.writeOutput(NeoJspWriter.java:228)
        at coldfusion.runtime.NeoJspWriter.flush(NeoJspWriter.java:362)
        at coldfusion.cfc.CFCProxy.flush(CFCProxy.java:357)
        at coldfusion.cfc.CFCProxy.doInvoke(CFCProxy.java:332)
        at coldfusion.cfc.CFCProxy.invoke(CFCProxy.java:302)
        at coldfusion.runtime.AppEventInvoker.onSessionEnd(AppEventInvoker.java:462)
        at coldfusion.runtime.SessionTracker.invokeOnSessinEnd(SessionTracker.java:292)
        at coldfusion.runtime.SessionTracker.cleanUp(SessionTracker.java:275)
        at coldfusion.runtime.SessionTracker.access$000(SessionTracker.java:43)
        at coldfusion.runtime.SessionTracker$SessionCleanUpAgent.run(SessionTracker.java:484)
        at coldfusion.scheduling.ThreadPool.run(ThreadPool.java:260)
        at coldfusion.scheduling.WorkerThread.run(WorkerThread.java:71)
 
The application.log:
 
"Error","scheduler-2","08/03/17","05:11:14","","'' The specific sequence of files included or processed is: '''' "
"Error","scheduler-2","08/03/17","05:11:14","","'' The specific sequence of files included or processed is: '''' "
"Error","scheduler-2","08/03/17","05:11:14","","'' The specific sequence of files included or processed is: '''' "
"Error","scheduler-2","08/03/17","05:11:14","","'' The specific sequence of files included or processed is: '''' "
"Error","scheduler-2","08/03/17","05:11:14","","'' The specific sequence of files included or processed is: '''' "
"Error","scheduler-2","08/03/17","05:11:14","","'' The specific sequence of files included or processed is: '''' "
"Error","scheduler-2","08/03/17","05:11:14","","'' The specific sequence of files included or processed is: '''' "
...
 
Please investigate CF team.  What is going on here?  Why are we getting these exceptions?   What damage are they causing?
 
CF2016, RedHat7, Apache 2.4, everything patched and current.
Steps to Reproduce:

Actual Result:

Expected Result:

Any Workarounds:

Attachments:

Comments:

Continuing today: 38 of them in a row: "Error","scheduler-2","08/04/17","17:51:21","","'' The specific sequence of files included or processed is: '''' " java.lang.NullPointerException at coldfusion.runtime.NeoJspWriter.writeOutput(NeoJspWriter.java:228) at coldfusion.runtime.NeoJspWriter.flush(NeoJspWriter.java:362) at coldfusion.cfc.CFCProxy.flush(CFCProxy.java:357) at coldfusion.cfc.CFCProxy.doInvoke(CFCProxy.java:332) at coldfusion.cfc.CFCProxy.invoke(CFCProxy.java:302) at coldfusion.runtime.AppEventInvoker.onSessionEnd(AppEventInvoker.java:462) at coldfusion.runtime.SessionTracker.invokeOnSessinEnd(SessionTracker.java:292) at coldfusion.runtime.SessionTracker.cleanUp(SessionTracker.java:275) at coldfusion.runtime.SessionTracker.access$000(SessionTracker.java:43) at coldfusion.runtime.SessionTracker$SessionCleanUpAgent.run(SessionTracker.java:484) at coldfusion.scheduling.ThreadPool.run(ThreadPool.java:260) at coldfusion.scheduling.WorkerThread.run(WorkerThread.java:71) "Error","scheduler-2","08/04/17","17:51:21","","'' The specific sequence of files included or processed is: '''' " java.lang.NullPointerException at coldfusion.runtime.NeoJspWriter.writeOutput(NeoJspWriter.java:228) at coldfusion.runtime.NeoJspWriter.flush(NeoJspWriter.java:362) at coldfusion.cfc.CFCProxy.flush(CFCProxy.java:357) at coldfusion.cfc.CFCProxy.doInvoke(CFCProxy.java:332) at coldfusion.cfc.CFCProxy.invoke(CFCProxy.java:302) at coldfusion.runtime.AppEventInvoker.onSessionEnd(AppEventInvoker.java:462) at coldfusion.runtime.SessionTracker.invokeOnSessinEnd(SessionTracker.java:292) at coldfusion.runtime.SessionTracker.cleanUp(SessionTracker.java:275) at coldfusion.runtime.SessionTracker.access$000(SessionTracker.java:43) at coldfusion.runtime.SessionTracker$SessionCleanUpAgent.run(SessionTracker.java:484) at coldfusion.scheduling.ThreadPool.run(ThreadPool.java:260) at coldfusion.scheduling.WorkerThread.run(WorkerThread.java:71)
Comment by Terry F.
460 | August 05, 2017 12:00:31 AM GMT
A couple of other people have reported this problem in the past without any resolution: https://forums.adobe.com/thread/2245288
Comment by Terry F.
461 | August 05, 2017 12:12:52 AM GMT
It seems as if ColdFusion is passing a null Session or Application scope object to the onSessionEnd method.
Vote by A. B.
466 | August 06, 2017 12:00:29 PM GMT
Terry, I do not see those errors in my CF exception log file with a simple CF application using on onSession start/end methods... I'm using CF2016 u4 in Windows. Perhaps you can share some more details. is there a pattern to when that happens, like only when the server is under load. Can you share some more details on the nature of your application.. and may be the what you are doing in your Application cfc. You may mail the relevant part of your code to me at pnayak@adobe.com
Comment by Piyush K.
462 | August 11, 2017 08:26:39 AM GMT
Hi Piyush -- There was no notable pattern to these errors. They occurred once a day on average, at random times, often in periods of time with light traffic. Our server is pretty heavily trafficked overall. I think I have found the root of the problem, although not sure why.... I noted in the logs that these errors started occurring after we modified one or two scheduled tasks in the scheduled task administration section. So, my theory is that some problem with our neo-cron.xml file (created by the administrator) set off some chain of events that messed up sessions. I loaded back up an old backup of neo-cron.xml and these errors appear to have stopped occurring. If the server runs for another day or two without these errors popping up, I will email you the two neo-cron.xmls so you can compare them and see what might be going on. Thanks
Comment by Terry F.
463 | August 18, 2017 05:14:48 PM GMT
Terry, Sounds good. So, have we seen any of those errors yet? You must have tried a text content comparison of the 2 cron config files. How were they different?
Comment by Piyush K.
464 | August 23, 2017 08:38:25 AM GMT
Hi! Changing back to the old neo-cron does appear to have solved MOST of the problem. I did get ONE set of the errors above after reverting the neo-cron, but it has stopped happening daily. I will email you the two neo-cron files to compare and test with. I was unable to determine the differences because linux diff considers the entire file a single line.
Comment by Terry F.
465 | August 31, 2017 06:35:45 PM GMT