Return-Path: Mailing-List: contact commons-user-help@jakarta.apache.org; run by ezmlm Delivered-To: mailing list commons-user@jakarta.apache.org Received: (qmail 29567 invoked from network); 6 Mar 2003 04:15:07 -0000 Received: from ext-nj2gw-3.online-age.net (216.35.73.165) by daedalus.apache.org with SMTP; 6 Mar 2003 04:15:07 -0000 Received: from int-nj2gw-2.online-age.net (int-nj2gw-2.online-age.net [3.159.236.66]) by ext-nj2gw-3.online-age.net (8.12.8/8.12.8/990426-RLH) with ESMTP id h264FFAt018293 for ; Wed, 5 Mar 2003 23:15:15 -0500 (EST) Received: from kans015cercge.erc.gecapital.com (localhost [127.0.0.1]) by int-nj2gw-2.online-age.net (8.12.3/8.12.3/990426-RLH) with ESMTP id h264F9YV018118 for ; Wed, 5 Mar 2003 23:15:09 -0500 (EST) Received: by ercmailhost.erc.gecapital.com with Internet Mail Service (5.5.2656.59) id ; Wed, 5 Mar 2003 22:15:08 -0600 Message-ID: From: Gaurav.Goyal@ercgroup.com To: commons-user@jakarta.apache.org Subject: RE: [DBCP] Exceptions in multi user environment. Date: Wed, 5 Mar 2003 22:15:07 -0600 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2656.59) Content-Type: text/plain; charset="iso-8859-1" X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N Is it possible that in some situation causes the pool to store two references to the same connection? -----Original Message----- From: Rodney Waldhoff [mailto:rwaldhoff@apache.org] Sent: Monday, March 03, 2003 5:21 PM To: Jakarta Commons Users List Subject: RE: [DBCP] Exceptions in multi user environment. On Mon, 3 Mar 2003, srinath narasimhan wrote: > I am using DBCP and found that similar error happens if one of my thread > crashes and does not return > the connection to the pool. I beleive that the connection is closed, but it > is not removed from the pool > for some reason. Once I fix the thread crash issue, it seems to go away. > But I was able to reproduce the problem for certain. The pool should see the connection as "active" in this case, and not return it to the client code. (Assuming the "abandonded object" stuff isn't configured.) Indeed, it is likely the pool doesn't even have a reference to the object in this case, which would allow the gc process to close the connection, but wouldn't cause it to be returned by the pool again. > > The way I found is, I had debug in my object shutdowns and printed the > thread which is executing > the shutdown, I found that whenever Finalizer thread was executing through > the shutdown ( which happens during crash ) > I got this error on the next database access. > > Hope this helps. > --------------------------------------------------------------------- To unsubscribe, e-mail: commons-user-unsubscribe@jakarta.apache.org For additional commands, e-mail: commons-user-help@jakarta.apache.org