Interesting... I'm not using CFHEAD at all... But I would suspect that both CFLOCATION and CFAPPLICATION (setting cookies) use the same functionality internally. > As for the cfid/app unique combination error, you mentioned that > the errors > shows up on framed cfm pages? Is it still true for a non-framed page? I thought that made a difference, but the error is actually displaying on the Framset page... The individual frame pages never load because of the error and/or lack of correct headers I suppose. I am pretty sure that the use of frames doesn't have anything to do with this problem. Your info is helpful, but I still can't see the reason for this error, or the long term solution... I'm continuing to tinker with the code to try to determine what will make it work without error. I'm kinda thinking that the first template is not finished saving the client data when the second page is called. The second page then thinks that the client var pair does not exist in the database, and tries to create it. But then why only on IE 5.5? It's fine on IE 4, and NS. Maybe IE 5.5 processes location headers faster? Fast enough to cause this problem? Still in the dark... -Cameron -------------------- Cameron Childress ElliptIQ Inc. p.770.460.7277.232 f.770.460.0963 > -----Original Message----- > From: Xing Li [mailto:[EMAIL PROTECTED]] > Sent: Monday, November 13, 2000 6:12 PM > To: CF-Talk > Subject: Re: Duplicate key row in object 'CDATA'? > > > Just checked the data scheme for CDATA and the "id1" is an unique index of > the two columns "cfid" and "app". I have never experienced the culmulative > error as you have described but I have gotten cflocation errors where html > headers show up on only IE browsers. Many this will help a bit. > > If you use a wrapper for <CFHEAD> to cfhead a bunch of generated > html to the > header area and do a cflocate to another page, IE will, not > always, but will > frequently generate a type of error where you either a blank page > or a full > page with some html header shown at the top. When the html headers are > shown, it seems like the cfserver has started but aborted the first set of > html headers and then continued with a second set of the same > headers. This > eror is much more severe for Mac IE but is part of Win IE as well. I don't > think it's a browser issue but rather the cfserver's way of handling the > cfhead operation. Netscape does not have this problem which is exactly as > you have discribed so this two problems might be related. > > As for the cfid/app unique combination error, you mentioned that > the errors > shows up on framed cfm pages? Is it still true for a non-framed page? > > Xing > > > ----- Original Message ----- > From: "Cameron Childress" <[EMAIL PROTECTED]> > To: "CF-Talk" <[EMAIL PROTECTED]> > Sent: Sunday, October 29, 2000 1:37 PM > Subject: RE: Duplicate key row in object 'CDATA'? > > > > > I've started to get the following error in my app: > > > > > > --- > > > [Microsoft][ODBC SQL Server Driver][SQL Server]Cannot insert duplicate > key > > > row in object 'CDATA' with unique index 'id1'. > > > > As a follow up to my original post, I have found out a great deal more > about > > the problem, though I still don't have a solution. Here's a post I just > > made to the Forums about it. > > > > Common CF-Talk, don't fail me now.... > > > > -- > > I've having the same error. Here's what info I can > contribute... I am able > > to reproduce the error intermittantly using IE 5.5, but not > with Netscape > > 4.7 (same client machine). I also have discovered a bit of code which I > can > > comment out that makes the problem completely do away, but I > have no idea > > why. > > > > The particular place in my app where this error occurs is when a user is > > logging into my app. After the login is authenticated via a > CFSTOREDPROC, > a > > whole slew of client vars are set, including one which is a WDDX packet > (in > > case that matters). > > > > After that, a quick CFIF test is done on the resultsets of the > STOREDPROC. > > No client vars are accessed during the execution of this CFIF, > nor inside > it > > if it returns true. If the CFIF does return true, an email is generated > > using CFMAIL and more of the STOREDPROC resultsets. > > > > That CFIF is the code I can comment out to make it all work. I > also tried > > commenting out the CFMAIL (so that nothing happened if the CFIF returns > > true), and I can reproduce the error. > > > > Also, if I replace the CFIF with a CFOUTPUT which output the same var > called > > in the CFIF, I can also generate the error. > > > > But yes, there's more! After the CFIF statement, the user is > CFLOCATIONed > to > > the entrance page to the app. The CFLOCATION does use > ADDTOKEN="Yes", but > > deleting that does not prevent the error from happening. > > > > Funny thing is, that the error is not happening on the page > with the CFIF, > > it's happening on the entry page to the app. Something in the login > template > > (cfm page 1) is causing the entry page (cfm page 2) to fail. > > > > The offending sequence seems to be (1) call a storedproc (it returns > several > > recordsets) (2) set a bunch or client vars using that data (3) call data > > from the storedproc recordset again. (4) CFLOCATION to another page. > > > > One other note here in ramblings... When you view the souce of the page, > it > > contains the error, then a bunch of raw HTML headers (which should not > show > > in the browser), and then the CFM page, correctly executed! > > > > So, my conclusion is... Well, I dunno really, but I can reproduce the > error, > > and I *can* make it go away (in my app). I just don't know why. > > > > Also, I wonder why it's only happening in IE,a nd not NS? Maybe > something > to > > do with the way each browser executes the CFLOCATION? Maybe the way each > > stores the CFID/CFTOKEN cookies? > > > > Anyone have any ideas? > > > > -Cameron > > > > -------------------- > > Cameron Childress > > ElliptIQ Inc. > > p.770.460.7277.232 > > f.770.460.0963 > > > > > -------------------------------------------------------------------------- > ---------------------- > > Archives: http://www.mail-archive.com/[email protected]/ > > Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists or send a message with 'unsubscribe' in the body to [EMAIL PROTECTED] > ---------------------------------------------------------------------------- -------------------- Archives: http://www.mail-archive.com/[email protected]/ Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists or send a message with 'unsubscribe' in the body to [EMAIL PROTECTED] ------------------------------------------------------------------------------------------------ Archives: http://www.mail-archive.com/[email protected]/ Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists or send a message with 'unsubscribe' in the body to [EMAIL PROTECTED]
