[ 
https://issues.apache.org/jira/browse/VELOCITY-719?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12711720#action_12711720
 ] 

Nathan Bubna commented on VELOCITY-719:
---------------------------------------

This exception is secondary.  It looks to me like the log initialization threw 
an exception that prevented resourceManager initialization, which they gave you 
this NPE when you tried to access a resource.   Do you have the exception/stack 
from the log's failure to initialize?   I don't see any way for the 
resourceManager to be null (as your exception indicates), unless 
initializeLog() threw an exception and you swallowed it and continued on.

We could consider changing the default value of runtime.log to blank in 2.0, 
but not before due to the need for backward compatibility.  I'd also quibble 
that vast majority of those complaints about velocity.log are from pre-1.5 
versions of Velocity.  I've put extensive work into improving the log 
facilities in the last 4 years.

> Unwritable velocity.log appears to cause Velocity initialization to fail, 
> leaving velocity dead and useless. 
> -------------------------------------------------------------------------------------------------------------
>
>                 Key: VELOCITY-719
>                 URL: https://issues.apache.org/jira/browse/VELOCITY-719
>             Project: Velocity
>          Issue Type: Bug
>          Components: Engine
>    Affects Versions: 1.6.2
>            Reporter: Cott Lang
>
> When we upgraded to 1.6.2 from 1.5.0, we began experiencing failures:
> Caused by: java.lang.NullPointerException
>  at 
> org.apache.velocity.runtime.RuntimeInstance.getTemplate(RuntimeInstance.java:1400)
>  at 
> org.apache.velocity.runtime.RuntimeSingleton.getTemplate(RuntimeSingleton.java:326)
>  at org.apache.velocity.app.Velocity.mergeTemplate(Velocity.java:352)
> We eventually traced this back to Log4JChute attempting to create 
> velocity.log in an unwritable $CWD.
> I realize you may or may not consider this a bug, as you may wish to have the 
> system fail if it can't log.  However, I couldn't even find any output 
> indicating the initial failure to create velocity.log causing Velocity to 
> fail to initialize. :)
> On a related note, is it possible to change the runtime.log default to BLANK 
> so that velocity defaults to using a pre-existing log4j config if it exists?
> Although it's certainly an established tradition at this point, It seems a 
> little backward to default to creating it's own log4j configuration - and a 
> google search on velocity.log produces a very large number of people dealing 
> with the problem of Velocity trying to create a velocity.log somewhere that 
> doesn't work. :)

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


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to