Catastrophic Bug When Set DML_RETURNING_VALUE To "YES"

Hi everyone.
I have a serious problem with some forms that are built on tables having unique keys or primary keys, and the block DML_RETURNING_VALUE property set to "YES". The problem is when updating a record and set its unique key column value to a value already exists in other record, the connection to database is lost and I must reopen the form again. I will show you the steps I went through to build this (and other) form and to show this error.
1- Build a table with unique key column.
2- Fill the table with some data (more than one record).
3- Build a form with one base block on this table, and set its DML_RETURNING_VALUE property to "YES".
4- Run the form and query the previously inserted records.
5- Try to update one of the records by setting its unique key field to a value previously entered in other record and press "Save".
6- The message "FRM-40509: ORACLE error: unable to UPDATE record" appears.
7- Set the unique key field back to its original value and press "Save".
8- The message " FRM-40400: Transaction complete: 1 records applied and saved" appears.
9- Now, set the value again to a previously entered value in other record and press "Save".
10- The message "FRM-40509: ORACLE error: unable to UPDATE record" appears.
11- Again, set the field value back to its original state and press "Save".
12- The message "FRM-40509: ORACLE error: unable to UPDATE record" appears, and it is followed by a second message " FRM-40655: SQL error forced rollback: clear form and re-enter transaction". From now on, you cannot do anything on the form because the connection to the database is lost.
When I set DML_RETURNING_VALUE property to "No", none of these problems exists, and I can update the record with wrong or right values (of course with error messages in case of wrong values) many times without any connection loss.
I wonder why this behavior exists. What should I do to avoid this problem?
I found that DML_RETURNING_VALUE property is a design time property and I cannot set it to any value programmatically (in pre-form trigger for example).
Please note that I am using Oracle Designer6i to generate the forms, and Forms6i runtime with patch 9. I tried to run this form against Oracle database 8i and 9i, but the same problem exists for both databases. If I use Forms6i to generate the forms, it will be no big deal as I will set this property to No, but the problem is when using Designer6i to generate the forms, Designer always set this property to YES, and it will be tedious and time consuming job if I must open every FMB file and set this property to NO for all blocks in the form.
Thank you in advance.

You are using a very old version of Forms 6i. If you upgrade to Patch 17, which is the most recent and last patch, you probably won't have the problem any more.
That is the first thing Oracle would tell you to do if you opened a TAR.

Similar Messages

Random text bug when set-top is in the off/standby mode.

Anyone else every see this when set-top is in standby with the TV still on? QIP6416-2 running 1.6.2 of the IMG in VHO8.

The ipad may keep a connection alive in sleep mode.. but it should only use local connection AFAIK, and no internet content.
I have a similar setup with iPad and although it has an IP it is not actually reachable on the network. I will track usage and see if it is actually using anything but I would doubt it.
Now of course it may depend on what apps you have running. I really doubt the ipad will be using that much of your download unless you are watching movies on it.

[svn:fx-trunk] 9284: Fix bug: Spark BasicLayout should throw error/ RTE when setting useVirtualLayout

Revision: 9284
Author:   [email protected]
Date:     2009-08-13 15:33:07 -0700 (Thu, 13 Aug 2009)
Log Message:
Fix bug: Spark BasicLayout should throw error/RTE when setting useVirtualLayout
QE notes: None
Doc notes: None
Bugs: SDK-22211
Reviewer: Hans
Tests run: checkintests, gumbo/core/Group, gumbo/core/DataGroup, gumbo/core/List
Is noteworthy for integration: No
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-22211
Modified Paths:
    flex/sdk/trunk/frameworks/projects/spark/bundles/en_US/layout.properties
    flex/sdk/trunk/frameworks/projects/spark/src/spark/layouts/BasicLayout.as

This bug figures out also when creating a custom spark ComboBox, then trying to programatically update the userProposedSelectedIndex property. The proposed selected index is selected, but does not apply the same skin as when mouse is on rollover or item is selected due to up and down keys.
The issue seems like updating the status of the item renderer to rollover or selected to get the same skin applied.
Please could you attach DropDow nList.as that you edited ?
Thank you so much.

[svn:cairngorm3:] 17872: Popup: Exposed set accessor of open for view state usage and fixed bug when closing from model within popup .

Revision: 17872
Revision: 17872
Author:   [email protected]
Date:     2010-09-27 08:04:40 -0700 (Mon, 27 Sep 2010)
Log Message:
Popup: Exposed set accessor of open for view state usage and fixed bug when closing from model within popup. http://www.spicefactory.org/forum/viewtopic.php?p=3932. Added new sample to show model and view state interactions.
Modified Paths:
    cairngorm3/trunk/libraries/Popup/src/com/adobe/cairngorm/popup/PopUpBase.as
    cairngorm3/trunk/libraries/Popup/src/com/adobe/cairngorm/popup/PopUpFactory.as
    cairngorm3/trunk/libraries/Popup/src/com/adobe/cairngorm/popup/PopUpWrapper.as
    cairngorm3/trunk/libraries/PopupTest/.actionScriptProperties
    cairngorm3/trunk/libraries/PopupTest/src/samples/MyPopup.mxml
    cairngorm3/trunk/libraries/PopupTest/src/samples/PopUpPM.as
Added Paths:
    cairngorm3/trunk/libraries/PopupTest/src/PopUpModelExample.mxml
    cairngorm3/trunk/libraries/PopupTest/src/samples/MyPopupWithModel.mxml

Lots to be excited about. BUT...Just updated to 8.1 on one of my computers to test it out...
Sadly, the following issue is NOT fixed for me. Is it with new projects only? I haven't started a new project on 8.1 I've only opened an old project but the issue that I'm referring to which involves also a delay/freeze of anywhere from a few seconds to a MINUTE while the render bar goes from yellow to red and then back to yellow is still there. Boo.
On the list above...
Switching between sequences can turn the render bar red.
This issue is incredibly easy to replicate. I really do hope it gets fixed. Here are the steps:
1) Create two sequences with multiple short clips (the more clips the better)
2) Make sure GPU acceleration is enabled.
3) Add warp stabilizer to the clips in both sequences. Again, the more clips the longer the delay/freeze will be as the render bar goes from YELLOW to RED and then eventually back to YELLOW again.
4) Hit SAVE.
5) Now, toggle between sequences. You'll hit a short delay with a few clips stabilized and a LONG delay with lots stabilized. In my case this delay is around a full minute. The render bar will go from YELLOW to RED and then back to yellow...eventually.
6) Toggle back to the original sequence and the delay occurs again.
7) Once you've toggled between sequences and have gone through this delay, that's it, there is no longer a delay...UNTIL...(and here's the big kicker)...until the project is SAVED again. After that the issue returns when you toggle between sequences. Both saving manually and AUTOSAVING cause this issue. No way around it except not editing with GPU acceleration.
Not sure what the issue is...caching issue when saving maybe? Either way, it stinks and it's still there in 8.1. Big bummer.
Again, lots to be excited about with this release but I really was hopefully this specific issue was resolved.
Sigh.

Bug in FB 4.7 when setting verbose-stacktraces to True

Hello guys,
When setting the verbose-stacktraces to true without setting the debug parameter to true, it has no effect. When I get a stacktrace it doesn't include any line numbers as stated on the documentation. I'm currently using Flash Builder 4.7, Adobe Flex 4.6 and Air 4 SDK to develop my mobile application. Also, I'm using the compiler options: -locale en_US -swf-version=23 -verbose-stacktraces=true.
I need to be able to see the line numbers so I would really appreciate any help on this issue.
Regards,
Sebastian

I also tried with replacing -verbose-stacktraces=true with -compiler.verbose-stacktraces=true as I've seen in some examples but nothing happens.
Next you have an example of a stacktrace missing the line numbers:
TypeError: Error #1009
        at components.visual::BaseECView/configureComponents()
        at views.home::Home/createChildren()
        at mx.core::UIComponent/initialize()
        at spark.components::View/initialize()
        at mx.core::UIComponent/http://www.adobe.com/2006/flex/mx/internal::childAdded()
        at mx.core::UIComponent/addChildAt()
        at spark.components::Group/addDisplayObjectToDisplayList()
        at spark.components::Group/http://www.adobe.com/2006/flex/mx/internal::elementAdded()
        at spark.components::Group/addElementAt()
        at spark.components::Group/addElement()
        at spark.components::SkinnableContainer/addElement()
        at spark.components::ViewNavigator/createViewInstance()
        at spark.components::ViewNavigator/commitNavigatorAction()
        at spark.components::ViewNavigator/commitProperties()
        at mx.core::UIComponent/validateProperties()
        at mx.managers::LayoutManager/validateProperties()
        at mx.managers::LayoutManager/doPhasedInstantiation()
        at mx.managers::LayoutManager/doPhasedInstantiationCallback()
        at flash.utils::Timer/tick()

HT5622 When setting up an apple ID for a child in family sharing, the verification questions appear in French, even though the iPhone setting is in English...

Possible bug? Setting up apple id for child, under icloud, family sharing, and the 3 questions come up in Frnech, even though the phone is set to English and the remainder of screens have all been in English,
Is this a bug or is there a way to display the questions in English please,
thanks

I agree.
I found this 'bug' out the other day when getting ready for my new iPhone I was setting up my iPad with some new apps too and wanted to work with Find My iPhone.
Here's the deal; my main AppleID is just a simple 'Username' not a '[email protected]' AppleID.
I've had this since the need for AppleIDs were used for these forums and the iTunes Store, AppStore, etc.
So it works fine and well with that stuff but not with find my iphone on my iPad as it wants a full email account, either a '@me.com' or a '[email protected]' AppleID which I did create but I think it was used as Job profile or something other along those lines (NOT used for a MobileMe account or .mac either).
And I only have access to one email now and I dont feel comfortable having multiple email addresses that I dont check or have no use for, call my simple or naive but thats just how it goes.
What I'd like to do is be able to use either AppleID for my Find My iPhone app and everything else.
And I think thats a good point, '@me.com's for all. make it a whole lot easier in my opinion.
merge all the apple accounts you created into a MobileMe, simplifies the situation and makes it easier connect to all Apple Services.
My itunes account, apple.com store account, appstore account, main AppleID for the past couple years it "Maxwell Stanley" but it does NOT work with Find My iPhone. So i created [email protected] (example.com is in place of my true email domain for privacy reasons) and that doesnt work as it statse it Cannot Be used.
My 'Maxwell Stanley" Apple ID's email associated with it was [email protected] i tried changing it to [email protected] but still doesnt recognize that as an AppleID even though its associated with "Maxwell Stanley".
I event ried to use the SAME email that is associated with my 2nd AppleID '[email protected]' with the appleid 'Maxwell Stanley' and no dice.
running out of options it seems.
Any idea on the best solution?
Thank you!

Bug when using IBOT

Hi all,
has anyone had this kind of error?
I am using IBOT, and I set Schedule_ to Start Date=21/09/2009, Time of Day=08:00, Set Time Zone=GMT+01:01, and Recurrence_ to Every 1 week(s) on MON.
When I look overview of this IBOT, everything is fine, but when I go Show IBOTs acting on my behalf Next Run is not date that I had set. And IBOT is not running in time that he should. But when I set Recurrence_ Daily Every 7 day(s) every thing is OK. IBOT is running and Next Run shows correct date and time.
Thanks

This looks the same as bug 5539348 "Schedule an iBot to run immediately and weekly, the immediate run doesn't take place". According to that bug when you say start immediately and run every 1 week you are actually saying start the schedule now not start the ibot?! So a schedule is created in the scheduler to run your ibot every week on Mon but it does not run it immediately.
I think this is a bug with the weekly schedule though as setting it to 7 days seems to fix it. The bug database seems in infer that later releases of 10g fixed this problem so it might be worth upgrading to a newer patch release if it becomes a problem and seeing if this fixes it.

Report Viewer renders incorrectly when set to "IMAGE"

I read this;
http://social.msdn.microsoft.com/Forums/en-US/b4a6eb43-0013-435f-9d11-00ee26a8d017/report-viewer-error-on-export-pdf-or-excel-from-azure-web-sites?forum=windowsazurewebsitespreview
And after upgrading to "Basic" mode on Azure websites I was able to render PDF, EXCEL and WORD formats. However when set to IMAGE the rendered binary is entirely black. No errors are thrown. The binary contains data (33kB). I
have tried changing the deviceInfo settings to include <OutputFormat>JPEG</OutputFormat> and a few other variants.
I have super verified that when run on my local machine the render works correctly but not on Azure websites.
I imagine that the class of bug we are dealing with here is a permutation of the API was accidentally not enabled/tested when making the fix for the GDI issue mentioned in the link above.
This capability is very useful for enabling MVC Restful API's solutions. I think the business value is very high. Please help, I figure it will be a quick fix?!

Thanks for the report. Can you supply me a sample piece of code that demonstrates the problem, or point me to your web site -> the best way is communicate that is using our new support site.
http://azure.microsoft.com/blog/2014/10/08/support-site-extension-for-azure-websites-preview-release/
Please include enough information so I can repro the scenario myself and see what failed.

[svn:fx-4.x] 15030: AIR Spark Window bugs when layoutDirection="rtl".

Revision: 15030
Revision: 15030
Author:   [email protected]
Date:     2010-03-25 11:49:00 -0700 (Thu, 25 Mar 2010)
Log Message:
AIR Spark Window bugs when layoutDirection="rtl". On Windows only, an empty window when created, didn't maximize or resize correctly. This class maintains its width/height in bounds rather than in the super class' _width and _height variables. Need to call setActualSize() whenever bounds is set so that the layoutWidth gets set correctly if this component is mirroring.
Fix an RTE in TitleWindow that I stumbled upon. Can not access the nativeWindow displayState if the window is closed.
QE notes:
Doc notes: None
Bugs: SDK-25978, SDK-25827, SDK-25825
Reviewed By: Jason
Tests run: checkintests
Is noteworthy for integration: No
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-25978
    http://bugs.adobe.com/jira/browse/SDK-25827
    http://bugs.adobe.com/jira/browse/SDK-25825
Modified Paths:
    flex/sdk/branches/4.x/frameworks/projects/airspark/src/spark/components/Window.as
    flex/sdk/branches/4.x/frameworks/projects/airspark/src/spark/components/windowClasses/Tit leBar.as

Stope
Mailing me ***
Sent from my I phone

[svn] 3791: Fix bug when resetting stylesheet on FxApplication

Revision: 3791
Author: [email protected]
Date: 2008-10-21 13:23:09 -0700 (Tue, 21 Oct 2008)
Log Message:
Fix bug when resetting stylesheet on FxApplication
SDK-17691 RTE loading styles
When the FxApplication replaces its style sheet, it unloads the previous skin and loads in a new one. With the new skin, it creates a new instance of the contentGroup. However, when the content is added to the contentGroup, it has already been parented by the previous version of the contentGroup. This causes the RTE because the content is not in the correct state.
The solution is to set the old contentGroup's content property to null when we remove the skin. We also force the contentGroup to validate immediately because the new contentGroup was processing its content array before the oldGroup had removed the content as children.
QE Notes: Needs testing
Doc Notes: none
Bugs: SDK-17691
Reviewer: Glenn
Ticket Links:
http://bugs.adobe.com/jira/browse/SDK-17691
http://bugs.adobe.com/jira/browse/SDK-17691
Modified Paths:
flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/FxContainer.as

The password history check is not honored in a password reset. It always been like that and it is documented:
"Password History Length Constraint: If all of the following conditions are true, the following constraints MUST be satisfied:
The requesting protocol message is a password change (as compared to a password set)."http://msdn.microsoft.com/en-us/library/cc245669.aspx
So changing it is not a correcting a bug but asking to change the design. From a pure support perspective, you switched from bug request to design change request. Which might still be arguable if you have a solid business case.
Password set is given through an extended right, so by default only administrators or operators have this right. So it comes down to trust those individual. If you cannot, then you don't give them the right and you implement other identity
management solution (Password Self Reset for example).
Other fancy and not useful information:
http://msdn.microsoft.com/en-us/library/cc245692.aspx
Note: Posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

When setting photo as lock screen/home screen wallpaper, quality is reduced

When setting a photo as the wallpaper for my lock screen/home screen the photo quality is dramatically reduced. This appears to be an issue since updating to iOS 8, as I have not changed my home screen wallpaper yet since updating, and when setting the same photo as my lock screen now, there is a noticeable difference. This only seems to be a problem when using landscape photo. Anyone else had this problem?

Hi Mara,
This is apparently a bug in iOS 7. Take a look at the posts to the right under "More Like This". Several users have come up with creative workarounds for this. Hopefully it will be fixed soon in an upcoming release!
Cheers,
GB

Weblogic bug when using setProperty?

          Hello,
          We're using Weblogic 5.1 SP8.
          The bug (I think it's a bug) occurs when setting a property to an empty value:
          the old value keeps remained.(see attachment)
          The problem occurs when the scope is set to session (using request works fine,
          application I didn't test).
          Will this issue being solved with installing SP9 (or 10?).
          Is this a known bug? If not where should I submit it (I took a look at the site
          but for now I didn't find a place where one can submit bugs)?
          All answers are welcome,
          Frederic Marius
          [att1.html]


          It's not a bug???
          So it is logic that when the property is made empty (when you have previously
          assigned it a value)
          - the property IS CHANGED when the bean is in the request scope
          - the property IS NOT CHANGED when scope is session
          This seems pretty confusing to me, and I can't imagine what could be the reason
          for wanting such a behaviour.
          When the spec says you shouldn't change the value of a property when it's according
          parameter has "" as value, that implies that you can only set values in a form
          and not unset them anymore. Seems pretty strange to me.
          Most logic behaviour is that the property is set to an empty string when you delete
          the values from the input field.
          =?iso-8859-1?Q?Anders_J=F6felt?= <[email protected]> wrote:
          >
          >
          >This is not a bug, according to jsp spec. 2.13.2.1 you can read about
          >=
          >the rules. 'If a parameter has the value of "" the corresponding =
          >property is not changed '
          >
          >/ Anders
          > "Frederic Marius" <[email protected]> wrote in message =
          >news:[email protected]...
          >
          > Hello,
          > We're using Weblogic 5.1 SP8.
          > The bug (I think it's a bug) occurs when setting a property to an =
          >empty value:
          > the old value keeps remained.(see attachment)
          > The problem occurs when the scope is set to session (using request
          >=
          >works fine,
          > application I didn't test).
          >
          > Will this issue being solved with installing SP9 (or 10?).
          > Is this a known bug? If not where should I submit it (I took a look
          >at =
          >the site
          > but for now I didn't find a place where one can submit bugs)?
          >
          > All answers are welcome,
          > Frederic Marius
          >
          >
          ><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
          ><HTML><HEAD>
          ><META content=3D"text/html; charset=3Diso-8859-1" =
          >http-equiv=3DContent-Type>
          ><META content=3D"MSHTML 5.00.2920.0" name=3DGENERATOR>
          ><STYLE></STYLE>
          ></HEAD>
          ><BODY bgColor=3D#ffffff>
          ><DIV><FONT face=3DArial size=3D2>This is not a bug, according to jsp
          >=
          >spec. 2.13.2.1=20
          >you can read about the rules. 'If a parameter has the value of ""
          >=
          >the=20
          >corresponding property is not changed '</FONT></DIV>
          ><DIV> </DIV>
          ><DIV><FONT face=3DArial size=3D2>/ Anders</FONT></DIV>
          ><BLOCKQUOTE=20
          >style=3D"BORDER-LEFT: #000000 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT:
          >=
          >0px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px">
          > <DIV>"Frederic Marius" <<A=20
          > =
          >href=3D"mailto:[email protected]">[email protected]=
          ></A>>=20
          > wrote in message <A=20
          > =
          >href=3D"news:[email protected]">news:[email protected]=
          >m</A>...</DIV><BR>Hello,<BR>We're=20
          > using Weblogic 5.1 SP8.<BR>The bug (I think it's a bug) occurs when
          >=
          >setting a=20
          > property to an empty value:<BR>the old value keeps remained.(see=20
          > attachment)<BR>The problem occurs when the scope is set to session
          >=
          >(using=20
          > request works fine,<BR>application I didn't test).<BR><BR>Will this
          >=
          >issue=20
          > being solved with installing SP9 (or 10?).<BR>Is this a known bug?
          >If =
          >not=20
          > where should I submit it (I took a look at the site<BR>but for now
          >I =
          >didn't=20
          > find a place where one can submit bugs)?<BR><BR>All answers are=20
          > welcome,<BR> Frederic Marius</BLOCKQUOTE></BODY></HTML>
          >
          >

Setting Printmotd to yes in sshd_config disallows SSH connections

Once I set Printmotd to yes, HUP the ssh daemon and try to establish a ssh connection, it just hangs indefinitely after accepting the password. This is on Solaris 9 with Open SSL. I have checked the following:
1. The sshd_config file is not the problem as it runs okay on another server without a problem.
2. Eliminated the possibility of a connection between the version of Solaris and the OpenSSL version by testing it on another server with the same versions.
3. Identified that only the Printmotd setting is causing new ssh connections to hang on these domains.
4. Checked that the sshd binary is also working fine by transporting it across to another test server.
5. Replicated the same settings as they exist on the other Solaris 9 server to these domains, however the ssh connections still failed.
6. Investigated the /etc/profile settings as well and they are the the same.
I have checked the permissions of /etc/issue and those are different. Can that be the cause of the problem? or is it something else that I have overlooked?

Using debug mode, when I try to login I get the completely same set of messages that i would get when ssh is working, except it hangs at the point indicated below:
debug1: next auth method to try is password
d36@nuspe's password:
debug1: ssh-userauth2 successfull: method password
debug1: channel 0: new [client-session]
debug1: send channel open 0
debug1: Entering interactive session.
debug1: client_init id 0 arg 0
debug1: channel request 0: shell
debug1: channel 0: open confirm rwindow 0 rmax 32768 -> Stops here
A successful ssh connection goes from the previous line to:
Last login: Mon Jun 15 13:38:48 2009 from nus007

[svn:fx-trunk] 8732: Do not need to invalidateDisplay list when setting either the horizontalScrollPosition or verticalScrollPostion since the

Revision: 8732
Author:   [email protected]
Date:     2009-07-22 08:00:53 -0700 (Wed, 22 Jul 2009)
Log Message:
Do not need to invalidateDisplay list when setting either the horizontalScrollPosition or verticalScrollPostion since the
ContainerController's setters call updateForScroll(), if scrolling needs to happen, which updates the container.
Also do not need to invalidate the display list when the scroller calls clipAndEnableScrolling install or uninstall the RET as a viewport. The scroller should do it if necessary.
QE notes:
Doc notes:
Bugs: SDK-22293
Reviewer: will be Gordon and Corey
Tests run: performance tests mentioned in the bug
Is noteworthy for integration:
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-22293
Modified Paths:
    flex/sdk/trunk/frameworks/projects/spark/src/spark/primitives/RichEditableText.as

The same id shows at the top of the Settings > iCloud screen on both your iPad and iPhone - if they are different then they are not being synced. And for contacts and calendars you need Settings > iCloud > Contacts and Settings > iCloud > Calendars 'on' on both devices

When setting up sync how does sync manage information from two computers? Does it merge the information?

When setting up sync how does sync manage information from two computers? Does it merge the information?

Hi!
Yes, the way Sync works is that it puts all the Bookmarks together and pushes them to all your devices. Same with History.
You will not loose any information in any of your computers.

Catastrophic Bug When Set DML_RETURNING_VALUE To "YES"

Similar Messages

Maybe you are looking for