[jboss-jira] [JBoss JIRA] (WFLY-3571) bad error message for invalid <resource-root> path in module.xml
Brian Stansberry (JIRA)
issues at jboss.org
Fri Jul 11 09:52:24 EDT 2014
[ https://issues.jboss.org/browse/WFLY-3571?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12984242#comment-12984242 ]
Brian Stansberry commented on WFLY-3571:
----------------------------------------
No worries about the description; I very much appreciate the report. I commented on that missing line just in case any of the other folks who are pretty familiar with this stuff happened to have a look at this, so they'd better understand.
It's trivial to have the stack trace always logged; it will take me about 1 min. The thing I'm thinking about is that a lot of times it's just noise, and we get lots of complaints about our noisy intimidating error messages. One heuristic I'm considering is checking if the caught exception returns null from getCause(). If so, don't log a stack trace. In the case of this module loading problem, there will be a cause. Basically the idea is the relevant exception type, OperationFailedException, is meant to provide the relevant data in its getFailureDescription message, and if the only stack trace is its own, it's not adding much. But if it includes a cause, then the stack trace of that cause is important and the whole stack should be logged.
I'm not convinced that's a good enough or complete enough solution though. There are some other cases I'm thinking about.
If we don't log the stack in some cases, we should go ahead and log it after the ERROR message as a separate DEBUG message.
> bad error message for invalid <resource-root> path in module.xml
> ----------------------------------------------------------------
>
> Key: WFLY-3571
> URL: https://issues.jboss.org/browse/WFLY-3571
> Project: WildFly
> Issue Type: Enhancement
> Security Level: Public(Everyone can see)
> Components: Domain Management
> Reporter: Karl Pietrzak
> Assignee: Brian Stansberry
> Priority: Minor
>
> h3. What
> If you have a typo in your path for your module.xml for a JDBC driver, e.g.,
> {code:xml}
> <?xml version="1.0" encoding="UTF-8"?>
> <module xmlns="urn:jboss:module:1.0" name="com.mysql">
> <resources>
> <resource-root path="DOES.NOT.EXIST.jar"/>
> </resources>
> <dependencies>
> <module name="javax.api"/>
> <module name="javax.transaction.api"/>
> </dependencies>
> </module>
> {code}
> WF just says:
> {noformat}
> 16:22:26,770 ERROR [org.jboss.as.controller.management-operation] (ServerService Thread Pool -- 30) JBAS014613: Operation ("add") failed - address: ([
> ("subsystem" => "datasources"),
> ("jdbc-driver" => "com.mysql")
> {noformat}
> h3. Why
> This is not very informative, unfortunately. Maybe it's logged as level DEBUG instead of WARN? If someone points me to the name of the class that handles this kind of thing, I can submit a patch/pull request.
--
This message was sent by Atlassian JIRA
(v6.2.6#6264)
More information about the jboss-jira
mailing list