The listener was hanging. A Tivoli alarm would have caught it, but I think that justs checks for the presense of a process (dosn't actually try to connect to the listener). The listener process was running, just not working. (Hopefully I can get the Tivoli people to cook up a better alarm trigger).
I have a feeling that the listener was giving ColdFusion enough to think it had a good connection. It was suggested that we put timeout attributes on our querys, but I don't think those are supported in the Oracle 8i driver we are using. I'm just suprised that the timeouts in the CF Admin didn't trigger. I'll check with the webmasters to ensure that they have optimal "connection timeout"/"login timeout" connections set for that datasource. Thanks :) Eric -----Original Message----- From: Dave Hannum [mailto:[EMAIL PROTECTED]] Sent: Thursday, November 29, 2001 8:43 AM To: CF-Talk Subject: Re: REPOST - cfquery timeout attribute Whenever I have an Oracle listner problem, try/catch always catches it. Are you sure it was a listner crash? Dave ----- Original Message ----- From: "Carlisle, Eric" <[EMAIL PROTECTED]> To: "CF-Talk" <[EMAIL PROTECTED]> Sent: Thursday, November 29, 2001 8:18 AM Subject: RE: REPOST - cfquery timeout attribute Every query I do is nested in a try/catch. To bad I can't timeout a try/catch block. :/ Thanks :) EC -----Original Message----- From: Dave Hannum [mailto:[EMAIL PROTECTED]] Sent: Thursday, November 29, 2001 8:17 AM To: CF-Talk Subject: Re: REPOST - cfquery timeout attribute Use CFTRY/CFCATCH around your queries and do a little custom error handling. It goes a LONG way! Dave ----- Original Message ----- From: "Carlisle, Eric" <[EMAIL PROTECTED]> To: "CF-Talk" <[EMAIL PROTECTED]> Sent: Thursday, November 29, 2001 8:04 AM Subject: REPOST - cfquery timeout attribute We had an Oracle listener hang. This caused big problems in ColdFusion. The "timeout" attribute just didn't seem to work. I posted this here before and it was suggested that the timeout clock doesn't start until the SQL is sent. Since the listener was hung, I'm not sure how far CF went with the connection. In anycase, the unresponsive threads brought down a handful of servers just because one datasource was inoperable. Is there anyway to avoid this in the future? It seems like CFs error correction (timeout in the CF admin) flew out the window. Can I get more error correction by using ADO? Thanks :) EC . ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Get the mailserver that powers this list at http://www.coolfusion.com FAQ: http://www.thenetprofits.co.uk/coldfusion/faq Archives: http://www.mail-archive.com/[email protected]/ Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists
