Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: ora-01003

Re: ora-01003

From: hpuxrac <johnbhurley_at_sbcglobal.net>
Date: 21 Mar 2007 11:53:59 -0700
Message-ID: <1174503239.364366.263720@y66g2000hsf.googlegroups.com>


On Mar 21, 2:37 pm, daniel.p.ngu..._at_gmail.com wrote:
> I was wondering if any one could extend their help. Thank you in
> advance for any one replying.
>
> Our .NET web application written in C#, running in IIS 6.0, pointing
> to Oracle 10. The web application utilize connection pulling. We are
> managing connection very tighly. We open the connection at the last
> minute, and dispose the connection right after use.
>
> We experience the "ora-01003" error randomly. After an IIS reset, the
> error does go away. The error comes back after some usage. The
> behavior is very random though.
>
> We've tried running differerent type of test against our application.
>
> We tried running an endurance test against our web application. On
> some test iteration, it might take about an hour. On other test
> iteration, it might take half an hour. This test make 1 single
> request only after the first request has finished.
>
> We also tried running a load test against our web application. Again,
> sometime it takes half an hour, others take about an hour. Sometime
> we cannot reproduce the error at all. This test creates simultaneous
> concurrent requests to our web application.
>
> Both test senerio, we cannot reproduce the error with confidence. The
> error happens very RANDOM. I've looked on the web for this sepecific
> oracle error and most people also experience the same behavior.
>
> --1--The error happens randomly only after some use.
> --2--Reseting the web server resolves the error.
> --3--The error comes back after some use.
>
> Most threads do not have a solution. Most threads suggest to reset
> the web server, which is not the answer we are looking for. Turning
> off connection pulling is also not an option in this senario.
>
> We are 99% pretty sure that oracle isn't the issue, because the
> version before this didn't experience this specific error.
>
> After spending the last few days debugging this, we "think" that the
> issue lies in the connection pool being corrupted. But we have no
> idea if that's the case.

After you working with microsoft on this problem? Received on Wed Mar 21 2007 - 13:53:59 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US