Another tracing bottleneck with commons-lang-2.0

[Ilkka Priha <im...@surfeu.fi>]

Hi

Here's one other trace call, which seems to cause each db query to take 
up to 10-30 secs with lang-2.0 instead of 100-200 msecs with lang-1.0.1.

The first one has been removed in cvs, thanks.

-- Ilkka

PS. We apply a customized OJB logger, which has been configured to 
filter out all debug level messages, but it still returned true for the 
isDebugEnabled(). That's why we saw only the reduced performance and not 
the large trace messages. Now that we've fixed our logger, the slowness 
is not a problem for us anymore.

/**
  * RsIterator constructor.
  * @param query the SELECT producing the underlying resultset
  * @param mif ClassDescriptor of the result-class
  * @param broker the broker we should use.
  */
public RsIterator(Query query, ClassDescriptor mif,
     PersistenceBrokerImpl broker)
{
     this(mif, broker);
     if (logger.isDebugEnabled())
--->    logger.debug("RsIterator(" + query + ", " + mif + ")");
     m_query = query;
     m_rsAndStmt = broker.serviceJdbcAccess().executeQuery(query, mif);
     prefetchRelationships(query);
}




---------------------------------------------------------------------
To unsubscribe, e-mail: ojb-dev-unsubscribe@db.apache.org
For additional commands, e-mail: ojb-dev-help@db.apache.org