cancel
Showing results for 
Search instead for 
Did you mean: 

Negative cache - ClassLoader

Former Member
0 Kudos

Hi,

I get this warning message when trying to access my webdynpro service :

com.company.ayantdoit.models.zayantdroitgetascmodel.Zayantdroit_S Found in negative cache -


Loader Info -


ClassLoader name: |com.company.tv1/essfrayantdroit3|

Parent loader name: |Frame ClassLoader|

References: common:service:http;service:servlet_jsp service:ejb common:service:iiop;service:naming;service:p4;service:ts service:jmsconnector library:jsse library:servlet common:library:IAIKSecurity;library:activation;library:mail;library:tcsecssl library:ejb20 library:j2eeca library:jms library:opensql common:library:com.sap.security.api.sda;library:com.sap.security.core.sda;library:security.class;library:webservices_lib;service:adminadapter;service:basicadmin;service:com.sap.security.core.ume.service;service:configuration;service:connector;service:dbpool;service:deploy;service:jmx;service:jmx_notification;service:keystore;service:security;service:userstore interface:resourcecontext_api interface:webservices interface:cross interface:ejbserialization sap.com/tcwddispwda sap.com/tcwdcorecomp service:webdynpro service:sld library:tcddicddicservices library:com.sap.aii.proxy.framework library:tcgraphicsigs library:com.sap.mw.jco library:com.sap.lcr.api.cimclient library:sapxmltoolkit library:com.sap.aii.util.rb library:com.sap.util.monitor.jarm library:tcddicddicruntime library:com.sap.aii.util.xml library:com.sap.aii.util.misc library:tc~cmi

Resources: /usr/sap/RW2/J00/j2ee/cluster/server0/apps/com.company.tv1/essfrayantdroit3/src.zip /usr/sap/RW2/J00/j2ee/cluster/server0/apps/com.company.tv1/essfrayantdroit3/webdynpro/public/lib/com.company.tv1essfr~ayantdroit3.jar

Loading model: {parent,references,local} -


Seems that it's a "negative cache" issue but I wonder how to solve it...

I found some posts about this: , and .

But I couldn't find my answer...

All I understand is that the "com.company.ayantdoit.models.zayantdroitgetascmodel.Zayantdroit_S" structure is defined twice on the server.

That would maybe be explained by the fact this DC has been duplicated in a new track, with the same name, and same package name. So we now have 2 DCs with the exact same name.

I tried undeploying this DC on the server (so it'd wipe everything out) and redeploying it with the latest version (the one from the new track) but it didn't change anything.

For information, I already tried to transport it again, but it didn't change anything.

Would someone have any idea ?

Edited by: Arnaud Leymet on Jun 30, 2008 1:14 PM

Accepted Solutions (0)

Answers (1)

Answers (1)

former_member190457
Contributor
0 Kudos

Hi,

You probably left out some runtime dependency: the engine cannot load your application without loading before the required dependecies and therefore stores your app id in the negative cache. Check the dependencies, restart your server to clear the negative cache and redeploy. The way to check your dependencies depends on how you are deploying your application (J2EE lib, WDP DC, ...).

Best regards

Vincenzo

Former Member
0 Kudos

Thank you Vincenzo for your quick reply.

We deploy our wdp applications over NWDI.

And this wdp project DC uses the default "Used DCs" (tc/logging, tc/cmi, com.sap.exception, etc.) plus a standard DC named com.sap.security.api.sda.

How can I check what/where my runtime dependencies are?

And when you refer to restarting the server, do you mean physically, only the nodes, or the J2ee instance...?

Regards

Former Member
0 Kudos

Well... our issue is repaired now.

Our problem was because of a Transport Request which wasn't present in this environment.

But for those who need to investigate the 'ClassLoader issue', you'll find some answers [here|http://help.sap.com/saphelp_nw04s/helpdata/en/25/724d457ea5ee469e35aa158d75c740/frameset.htm|ClassLoader Reference Viewer] and [here|http://help.sap.com/saphelp_nw04s/helpdata/en/25/724d457ea5ee469e35aa158d75c740/frameset.htm|Working with the Reference Viewer].

former_member187439
Active Participant
0 Kudos

Hi All, / Hi Arnaurd,

I'm facing the same problem. The models are in DC1.

DC2 uses DC1 for function modules and backend communication. After i re-imported one of the models in DC1, as i wanted to include a new FM, i by mistake released the activity of DC2 before releasing the activity of DC1. I guess this has caused the negative cache problem.

I restarted server, removed the entire configuration and imported newly, synced sourced, synced used DCs, rebuild, deployed in order.. tried watever i could. But the error still exists. Do you have any clue about the solution?

Thanks & Regards,

Kavitha