uima-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jerry Cwiklik (JIRA)" <uima-...@incubator.apache.org>
Subject [jira] Commented: (UIMA-1731) ResourceInitializationExceptions thrown by a deployed aggregate are only partially logged
Date Mon, 01 Feb 2010 18:48:20 GMT

    [ https://issues.apache.org/jira/browse/UIMA-1731?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12828217#action_12828217

Jerry Cwiklik commented on UIMA-1731:

I checked the code and discovered that the problem described here is due to a wrong style
of logging. In many places, while logging WARNING messages using these keys: 


the UIMA AS code uses something like this:

UIMAFramework.getLogger(CLASS_NAME).logrb(Level.WARNING, getClass().getName(),
                "takeAction", UIMAEE_Constants.JMS_LOG_RESOURCE_BUNDLE,
                "UIMAEE_exception__WARNING", new Object[] { e });

The above does not log the whole stack trace hiding a real cause. Change code to use this

UIMAFramework.getLogger(CLASS_NAME).logrb(Level.WARNING, getClass().getName(),
                "takeAction", UIMAEE_Constants.JMS_LOG_RESOURCE_BUNDLE,
                "UIMAEE_exception__WARNING", e);

The above is better than just using 

log(Level level, String aMessage, Throwable thrown);

since it identifies a method where the failure is detected.

> ResourceInitializationExceptions thrown by a deployed aggregate are only partially logged
> -----------------------------------------------------------------------------------------
>                 Key: UIMA-1731
>                 URL: https://issues.apache.org/jira/browse/UIMA-1731
>             Project: UIMA
>          Issue Type: Bug
>          Components: Async Scaleout
>            Reporter: Burn Lewis
>            Priority: Minor
>             Fix For: 2.3AS
> When the XcasCollectionReader fails to init because its input directory is missing it
reports the cause when run in core UIMA, i.e.
> Caused by: org.apache.uima.resource.ResourceInitializationException: Initialization of
annotator class "com.ibm.nlp.readers.XcasCollectionReader" failed.  (Descriptor: file:/.automount/hlthome01/a/hlthome01/homes/hlthome01.3/burn/workspace/factotem/desc/pipe/readers/XcasCollectionReader.xml)
> Caused by: org.apache.uima.resource.ResourceInitializationException: Invalid value for
parameter "InputDirectory" in component "null" -- directory "xmidata" does not exist.
> but when deployed under UIMA-AS this detailed exception is missing ... all we get is
the higher-level cause, e.g.
> 1/30/10 10:56:15 PM - 3: org.apache.uima.aae.controller.PrimitiveAnalysisEngineController_impl.initializeAnalysisEngine:
WARNING: org.apache.uima.resource.ResourceInitializationException: Initialization of annotator
class "com.ibm.nlp.readers.XcasCollectionReader" failed.  (Descriptor: file:/u/burn/workspace/factotem/desc/pipe/readers/XcasCollectionReader.xml)
> Also, the detailed message lists the component as "null" ... the collection reader follows
the example code and reports the error with:
>       throw new ResourceInitializationException(
>           ResourceConfigurationException.DIRECTORY_NOT_FOUND,
>           new Object[] { PARAM_INPUTDIR, this.getMetaData().getName(), directory.getPath()
> How should we retrieve the component name?

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message