Tomcat catalina.out file log error


#1

Running Lucee 5.29, Apache Tomcat/8.0.53 the catalina.out file in the /opt/tomcat322/logs is getting filled up with the following output:

t error.error_cfm$cf.udfCall(/error/error.cfm:120)
at lucee.runtime.type.UDFImpl.implementation(UDFImpl.java:107)
at lucee.runtime.type.UDFImpl._call(UDFImpl.java:357)
at lucee.runtime.type.UDFImpl.call(UDFImpl.java:226)
at lucee.runtime.type.scope.UndefinedImpl.call(UndefinedImpl.java:771)
at lucee.runtime.util.VariableUtilImpl.callFunctionWithoutNamedValues(VariableUtilImpl.java:756)
at lucee.runtime.PageContextImpl.getFunction(PageContextImpl.java:1718)
at error.error_cfm$cf.call(/error/error.cfm:17)
at lucee.runtime.PageContextImpl._doInclude(PageContextImpl.java:901)
at lucee.runtime.PageContextImpl._doInclude(PageContextImpl.java:823)
at lucee.runtime.PageContextImpl.doInclude(PageContextImpl.java:817)
at lucee.runtime.PageContextImpl.handlePageException(PageContextImpl.java:2032)
at lucee.runtime.util.PageContextUtil.getHandlePageException(PageContextUtil.java:232)
at lucee.runtime.exp.CatchBlockImpl.castToString(CatchBlockImpl.java:225)
at lucee.runtime.exp.CatchBlockImpl.castToString(CatchBlockImpl.java:217)
at lucee.runtime.op.Caster.toString(Caster.java:1858)

Running tail -f on catalina.out - this block gets repeated until the container runs out of virtual disk space.


within application.log in lucee/logs/

“Severity”,“ThreadID”,“Date”,“Time”,“Application”,“Message”

“ERROR”,“ajp-nio-9322-exec-5”,“10/10/2018”,“07:50:31”,"",";The key [LUCEE] does not exist, the structure is empty;lucee.runtime.exp.ExpressionException: The key [LUCEE] does not exist, the structure is empty

Any ideas of what could be causing this?

Thanks


#2

Need more information here. How are you deploying Lucee? As a WAR I’m assuming? Was the initial deployment successful? Are you attempting to execute custom code here?

The stack indicates Lucee is attempting to error, then errors while erroring which is causing the cascade. In order to determine the cause we need more information.


#3

We are deploying Lucee as a WAR via a custom deployment system written in Perl.
The initial deployment works to set up a Lucee server that seems fine otherwise.


#4

So can you get to your Lucee Admin URL’s okay after your initial WAR deployment?

Can you describe the configuration steps you take after you install the WAR?


#5

So can you get to your Lucee Admin URL’s okay after your initial WAR deployment?

* yes.

Can you describe the configuration steps you take after you install the WAR?

1.  Copy out config file containing our datasources for our Lucee application.

/WEB-INF/lucee-server/context/lucee-server.xml

  1. Create a symlink to the java keystone from the Lucee keystore.

     * we experienced issues when trying to make an CFHTTPS request to a web service prior to this step.
    
    1. Restart tomcat container. Restart Lucee

#6

Thank you for clarifying.

My guess would be that the issue is here. Instead of copying the entire file, try just replacing just the lines you need to replace. IE: the default datasource tag with your new datasource tags using a tried and true string replacement program such as “sed”.

If the file you are replacing the default file with contains unsupported characters, tags, or unsupported changes in the data, you could bork your Lucee instance - which it seems to be exactly what is happening.