Konrad Windszus created SLING-3459:
--------------------------------------
Summary: sling:call should not log exceptions with the full
stacktrace
Key: SLING-3459
URL: https://issues.apache.org/jira/browse/SLING-3459
Project: Sling
Issue Type: Bug
Components: Scripting
Affects Versions: Scripting JSP-Taglib 2.1.8
Reporter: Konrad Windszus
Currently within the sling:call tag all exceptions are both logged on error
level and rethrown
(https://fisheye6.atlassian.com/browse/~br=trunk/sling/trunk/bundles/scripting/jsp-taglib/src/main/java/org/apache/sling/scripting/jsp/taglib/CallTag.java?r=1398589&r=1520554&r=1398589#to139).
That is not a good practice, because the same stack traces would appear twice
in the log (once for the generic exception, logged by the CallTag and once for
the wrapped JspException).
Rather do not log the exception within the CallTag and leave that to other
handlers. Just rewrapping the exception into the JspException should be
enough. No information would be lost that way, and stack traces only logged
once in the log (by the code responsible to catch the JspException).
--
This message was sent by Atlassian JIRA
(v6.2#6252)
