SRTServletResponse Null Pointer Exception

bcny · Newbie Joined: Jul 14, 2005 Posts: 26

We are running WAS 6.0.2 (Released 7/12 - we upgraded because of this issue which still happens). When a url is loaded we keep recieving a null pointer exception thrown from the server itself. Sometimes the pages seems to work and other times they seem fail. Below is the stack trace, if any one could be of assistance it would be greatly appreciated.

[7/14/05 12:24:52:258 EDT] 000001f4 WebContainer E SRVE0232E: Internal Server Error. <br> Exception Message: [java.lang.NullPointerException
at com.ibm.ws.webcontainer.srt.SRTServletResponse.setDefaultResponseEncoding(SRTServletResponse.java(Compiled Code))
at com.ibm.ws.webcontainer.srt.SRTServletResponse.start(SRTServletResponse.java(Inlined Compiled Code))
at com.ibm.ws.webcontainer.srt.SRTConnectionContext.start(SRTConnectionContext.java(Inlined Compiled Code))
at com.ibm.ws.webcontainer.WebContainer.handleRequest(WebContainer.java(Compiled Code))
at com.ibm.ws.webcontainer.channel.WCChannelLink.ready(WCChannelLink.java(Compiled Code))
at com.ibm.ws.http.channel.inbound.impl.HttpInboundLink.handleDiscrimination(HttpInboundLink.java(Compiled Code))
at com.ibm.ws.http.channel.inbound.impl.HttpInboundLink.handleNewInformation(HttpInboundLink.java(Compiled Code))
at com.ibm.ws.http.channel.inbound.impl.HttpICLReadCallback.complete(HttpICLReadCallback.java(Compiled Code))
at com.ibm.ws.tcp.channel.impl.WorkQueueManager.requestComplete(WorkQueueManager.java(Compiled Code))
at com.ibm.ws.tcp.channel.impl.WorkQueueManager.attemptIO(WorkQueueManager.java(Compiled Code))
at com.ibm.ws.tcp.channel.impl.WorkQueueManager.workerRun(WorkQueueManager.java(Compiled Code))
at com.ibm.ws.tcp.channel.impl.WorkQueueManager$Worker.run(WorkQueueManager.java(Compiled Code))
at com.ibm.ws.util.ThreadPool$Worker.run(ThreadPool.java(Compiled Code))

bcny · Newbie Joined: Jul 14, 2005 Posts: 26

It seems to be throwing this error while trying to load an image. If we restart the server the image loads fine for a while, and then at some unconsistent point it starts throwing this exception trying to load that image. If you type the location of the image into a web browser you get a HTTP 500 Error. It doesn't make sense. Its not an image name problem because there are images in the same dir that have longer names and they all load properly and the other weird thing is restarting the server and the problem goes away (for a period of time).

Anyone have any suggestions?

roksmith · Newbie Joined: Nov 30, 2004 Posts: 20

make sure you don't have 2 applications with the same root context

bcny · Newbie Joined: Jul 14, 2005 Posts: 26

We don't, and as stated above, once the server is restarted it will act fine for a while and then start throwing exceptions again.

cog · Newbie Joined: Jul 26, 2005 Posts: 1

We have just noticed this same error. It is very specific to 1 image's name. If we change it by a single character, it works. Have you had any success in figuring this problem out yet?

bcny · Newbie Joined: Jul 14, 2005 Posts: 26

No we haven't, in fact now its happening on yet another image!

bradyp · Newbie Joined: Oct 04, 2005 Posts: 1

Sorry to be repetitive, but has anyone solved this problem? I am experiencing it problem now too. I haven't had much time to delve into it yet, but the apperence of the problem seems to have co-incided with the introduction of a new css stylesheet into our application which references images. Does this ring any bells?
Any help appreciated!

krausest · Newbie Joined: Oct 10, 2005 Posts: 1

I get the same error - and have no solution yet, but I'm in contact with IBM support.
Some things I still have to find out:
So far I've seen it only for new deployments (EAR files freshly deployed, no server restart), but I haven't tested thoroughly. Have you seen this exception also when you've stopped and restarted the app server after a fresh deployment?
Any other news on this issue?

was_vps · Newbie Joined: Aug 31, 2005 Posts: 1

Hi,

We opened a PMR on this, after many traces and mail to IBM support.
We found that the problem is related to the invocationCache.
The problem always occoured after a re-deploy,and this cache was not
cleaned up .
A workaround is to turn off this cahce. Setting invocationCacheSize=0

A fix are available soon

cheers, Bjørn

tomfordlsu · Newbie Joined: May 11, 2006 Posts: 1

We are seeing this problem also. IBM tells us to clear the wstemp directory but that is not a fix, it is a montonous work around. We are running 6.0.2 on Linux. Does anyone know if any of the existing IBM patches fix this particular problem?

I tried the invocationCacheSize=0 but for us it did not fix the problem.

bcny · Newbie Joined: Jul 14, 2005 Posts: 26

Sorry I thought I posted a reply that I got from was_vps. Here is the solution:

Servers > Application Servers > Server 1 > Java and Process Management > Process Definition > Java Virtual Machine > Custom Properties

Name = "invocationCacheSize"
Value = "0"
Description = ""

wenhuawu · Newbie Joined: Oct 25, 2006 Posts: 1

I following the following instruction and fixed my problem in RAD WAS test server.

-------------------------------------------------
Servers > Application Servers > Server 1 > Java and Process Management > Process Definition > Java Virtual Machine > Custom Properties

Name = "invocationCacheSize"
Value = "0"
Description = ""

jdm · Newbie Joined: Mar 07, 2007 Posts: 1

Hi,

i'm having the same problem on AIX and invocationCacheSize=0 did not make a difference. Has this been solved in a new release ?

jdm

Cynosure · Newbie Joined: Oct 30, 2008 Posts: 1

I am getting the same error in my environment any solution you have got. Help highly appreciated.