Thanks Dirk, I'm currently investigating an old mailet that I wrote back in 2002 as the culprit...given that noone else seems to be experiencing this problem.
Unfortunately, I'm not getting as much time as I'd like to work on this... I would assume that the problem isn't with your code right now. Thanks. -Chris > -----Original Message----- > From: Dirk Verbeeck [mailto:[EMAIL PROTECTED] > Sent: Wednesday, May 26, 2004 7:44 AM > To: [EMAIL PROTECTED] > Cc: Noel J. Bergman; 'James Developers List' > Subject: Re: Possible bug in RC3 with connection pool > > Attached you will find a org.apache.james.util.dbcp.JdbcDataSource > with more logging. > > Don't forget about the thread dumps when you get the failures. > > -- Dirk > > > Noel J. Bergman wrote: > > Dirk verbeeck wrote: > > > >>Chris Means wrote: > >> > >>>The update for DBCP didn't fix the problem. > > > > > > Bother ... worked all that time, and now failing again? > > > > > >>>I have restarted it, and turned on DEBUG for: spoolmanager, > >>>pop3server, smtpserver, and mailstore. > > > > > > That might not help, but we'll see. We'd want some detailed > > accounting from the connection pool when it is exhausted, > so that we > > could see how is holding the connections. > > > > > >>are the blocking threads (threaddumps?) If we suspect a leak then > >>ds.getNumActive() & ds.getNumIdle() should be printed (exception > >>logging in JdbcDataSource.getConnection). > >>If the numbers are correct (active=max and idle=0) then the DBCP > >>abandoned connection tracing should be enabled to trace where the > >>connection are allocated. > > > > > > FWIW, the James code that uses this is: > > > > > http://cvs.apache.org/viewcvs.cgi/james-server/src/java/org/apache/jam > > es/uti > > > l/dbcp/JdbcDataSource.java?annotate=1.1.2.3&only_with_tag=branch_2_1_f > > cs > > > > Any suggestions on instrumentation? A patch would be welcomed. At > > the moment, I'm rather rushed for time, and don't have time > to examine > > the issue in more detail immediately. We have our own > adapter for the > > BasicDataSource, so adding the instrumentation should be fairly > > straight foward. > > > > --- Noel > > > > > > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
