I have program that takes user and updates information about him/her in five tables. The process is fairly sophisticated as it takes many steps(pages) to complete. I have logs, sysout and syserr statements that helps me to find sql queries in IDE console but it doesn't have all of them. I've already spend many days to catch other missing queries by debugging but no luck so far. The reason why I am doing this is because I want to automate user information updates so I don't have to go through every page entering user details manually.
I wonder if I could just have some technique that will show me database table changes as I already know table names, by changes I mean whether it was update or insert statements and what exactly changed(column name and value inserted/updated). Any advice is greatly appreciated. I have IBM RAD and DB2 database. Thanks.
In DB2 you can track basic auditing information.
DB2 can track what data was modified, who modified the data, and the SQL operation that modified the data.
To track when data was modified, define your table as a system-period temporal table. The row-begin and row-end columns in the associated history table contain information about when data modifications occurred.
To track who and what SQL modified the data, you can use non-deterministic generated expression columns. These columns can contain values that are helpful for auditing purposes, such as the value of the CURRENT SQLID special register at the time that the data was modified. Possible values for non-deterministic generated expression columns are defined in the syntax for the CREATE TABLE and ALTER TABLE statements.
For example
CREATE TABLE TempTable (balance INT,
userId VARCHAR(100) GENERATED ALWAYS AS ( SESSION_USER ) ,
opCode CHAR(1)
GENERATED ALWAYS AS ( DATA CHANGE OPERATION )
... SYSTEM PERIOD (SYS_START, SYS_END));
The userId column stores who modified the data. This column is defined as a non-deterministic generated expression column that contains the value of SESSION_USER special register.
The opCode column stores the SQL operation that modified the data. This column is defined as a non-deterministic generated expression column and stores a value that indicates the type of SQL operation.
Suppose that you then use the following statements to create a history table for TempTable and to associate that history table with TempTable:
CREATE TABLE TempTable_HISTORY (balance INT, user_id VARCHAR(128) , op_code CHAR(1) ... );
ALTER TABLE TempTable ADD VERSIONING
USE HISTORY TABLE TempTable_HISTORY ON DELETE ADD EXTRA ROW;
Capturing SQL statements for a limited number of tables and a limited time - as far as I understand your problem - could be solved with the DB2 Audit facility.
create audit policy tabsql categories execute status both error type normal
audit <tabname> using policy tabsql
You have to have SECADM rights in theh database and the second command will start the audit process. You can stop it with
audit <tabname> remove policy
Check out the
db2audit
command to configure paths and extract the data from the audit file to a delimited file which then could be loaded again into the database.
The necessarfy tables can be created with the provided sqllib/misc/db2audit.ddl script. You will need the query the EXECUTE table for your SQL details
Please note that audit can capture huge amounts of data so make sure to switch it off again after you have catured the necessary information.
a problem has come up after a SQL DB I used was migrated to a new server. Now when trying to edit a record in Access (form or table), it says: WRITE CONFLICT: This record has been changed by another user since you started editing it...
Are there any non obvious reasons for this. There is noone else using the server, I've disabled any triggers on the Table. I've just found that it is something to do with NULLs as records that have none are ok, but some rows which have NULLs are not. Could it be to do with indexes? If it is relevant, I have recently started BULK uploading daily, rather than doing it one at a time using INSERT INTO from Access.
Possible problems:
1 Concurrent edits
A reason might be that the record in question has been opened in a form that you are editing. If you change the record programmatically during your editing session and then try to close the form (and thus try to save the record), access says that the record has been changed by someone else (of course it's you, but Access doesn't know).
Save the form before changing the record programmatically.
In the form:
'This saves the form's current record
Me.Dirty = False
'Now, make changes to the record programmatically
2 Missing primary key or timestamp
Make sure the SQL-Server table has a primary key as well as a timestamp (= rowversion) column.
The timestamp column helps Access to determine if the record has been edited since it was last selected. Access does this by inspecting all fields, if no timestamp is available. Maybe this does not work well with null entries if there is no timestamp column (see 3 Null bits issue).
The timestamp actually stores a row version number and not a time.
Don't forget to refresh the table link in access after adding a timestamp column, otherwise Access won't see it. (Note: Microsoft's Upsizing Wizard creates timestamp columns when converting Access tables to SQL-Server tables.)
3 Null bits issue
According to #AlbertD.Kallal this could be a null bits issue described here: KB280730 (last snapshot on WayBackMachine, the original article was deleted). If you are using bit fields, set their default value to 0 and replace any NULLs entered before by 0. I usually use a BIT DEFAULT 0 NOT NULL for Boolean fields as it most closely matches the idea of a Boolean.
The KB article says to use an *.adp instead of a *.mdb; however, Microsoft discontinued the support for Access Data Projects (ADP) in Access 2013.
Had this problem, same as the original poster. Even on edit directly using no form. The problem is on bit fields, If your field is Null, it converts Null to 0 when you access the record, then you make changes which this time is the 2nd change. So the 2 changes conflicts. I followed Olivier's suggestion:
"Make sure the table has a primary key as well as a timestamp column."
And it solved the problem.
I have seen a similar situation with MS Access 2003 (and prior) when linked to MS SQL Sever 2000 (and prior). In my case I found that the issue to be the bit fields in MS SQL Server database tables - bit fields do not allow null values. When I would add a record to a table linked via the MS Access 2003 the database window an error would be returned unless I specifically set the bit field to True or False. To remedy, I changed any MS SQL Server datatables so that any bit field defaulted to either 0 value or 1. Once I did that I was able to add/edit data to the linked table via MS Access.
I found the problem due to the conflict between Jet/Access boolean and SQL Server bit fields.
Described here under pitfall #4
https://blogs.office.com/2012/02/17/five-common-pitfalls-when-upgrading-access-to-sql-server/
I wrote an SQL script to alter all bit fields to NOT NULL and provide a default - zero in my case.
Just execute this in SQL Server Management Studio and paste the results into a fresh query window and run them - its hardly worth putting this in a cursor and executing it.
SELECT
'UPDATE [' + o.name + '] SET [' + c.name + '] = ISNULL([' + c.name + '], 0);' +
'ALTER TABLE [' + o.name + '] ALTER COLUMN [' + c.name + '] BIT NOT NULL;' +
'ALTER TABLE [' + o.name + '] ADD CONSTRAINT [DF_' + o.name + '_' + c.name + '] DEFAULT ((0)) FOR [' + c.name + ']'
FROM
sys.columns c
INNER JOIN sys.objects o
ON o.object_id = c.object_id
WHERE
c.system_type_id = 104
AND o.is_ms_shipped = 0;
This is a bug with Microsoft
To work around this problem, use one of the following methods:
Update the form that is based on the multi-table view
On the first occurrence of the error message that is mentioned in the "Symptoms" section, you must click either Copy to Clipboard or
Drop Changes in the Write Conflict dialog box. To avoid the repeated
occurrence of the error message that is mentioned in the "Symptoms"
section, you must update the recordset in the form before you edit
the same record again.
Notes
To update the form in Access 2003 or in Access 2002, click Refresh on the Records menu.
To update the form in Access 2007, click Refresh All in the Records group on the Home tab.
Use a main form with a linked subform
To avoid the repeated occurrence of the error message that is mentioned in the "Symptoms" section, you can use a main form with a
linked subform to enter data in the related tables. You can enter
records in both tables from one location without using a form that is
based on the multi-table view.
To create a main form with a linked subform, follow these steps:
Create a new form that is based on the related (child) table that is used in the multi-table view. Include the required fields
on the form.
Save the form, and then close the form.
Create a new form that is based on the primary table that is used in the multi-table view. Include the required fields on the
form.
In the Database window, add the form that you saved in step 2 to the main form.
This creates a subform.
Set the Link Child Fields property and the Link Master Fields property of the subform to the name of the field or fields that are
used to link the tables.
Methods from work around taken from microsoft support
I have experienced both of the causes detailed above: Directly changing data in a table that is currently bound to a form AND having a 'bit' type field in SQL Server that does not have the Default Value set to '0' (zero).
The only way I have been able to get around the latter issue is to add the default value of zero to the bit field AND run an update query to set all current values to zero.
In order to get around the former error, I have had to be inventive. Sometimes I can change the order of the VBA statements and move Refresh or Requery to a different location, thus preventing the error message. In most cases, however, what I do is DIM a String variable in the Subroutine where I call the direct table update. BEFORE I call the update, I set this String variable to the value of the Recordsource behind the bound form, thus capturing the exact SQL statement being used at the time. Then, I set the form's Recordsource to an empty string ("") in order to disconnect it from the data. Then, I perform the data update. Then, I set the form's Recordsource back to the value saved in the String variable, reestablishing the binding and allowing it to pick up the new value(s) in the table. If there is one or more subforms contained within this form, then the "Link" fields need to handled in a similar manner as the Recordsource. When the Recordsource is set to an empty string, you may see #Name in the now-unbound fields. What I do is simply set the Visible property to False at the highest possible level (Detail section, Subform, etc.) during the time when the Recordsource is empty, hiding the #Name values from the user. Setting the Recordsource to an empty string is my go-to solution when a coding change can't be found. I am wondering, though, if my design skills are lacking and there is a way to completely avoid the issue altogether?
One final thought on addressing the error message: Instead of calling a routine to directly update the data in the table table, I find a way to update the data via the form instead, by adding a bound control to the form and updating the data in that so that the form data and the table data do not become out of sync.
In order to get over this problem. I created VBA to change another field in the same row. So I created a separate field which adds 1 to the contents when I try to close the form. This solved the issue.
I've dealt with this issue with MS Access tables linked to MS SQL tables multiple times. The original poster's response was extremly helpful and was indeed the source of much of my issues.
I also ran into this issue when i accidently added a bit field with a space in the fieldname... yeah....
I had run alter table tablename add [fieldname ] bit default 0. i solution i found was to drop that field and not have a space in the name.
I had this issue and realized it was caused by adding a new bit field to an existing table. I deleted the new field and everything went back to working fine.
If you are using linked tables, ensure you have updated these and retry before doing anything else.
I thought I had updated them but hadn't, turns out someone had updated the form validation and SQL tables to allow 150 chars, but hadn't refreshed the linked table hence access only saw 50 char allowed - Boom Write conflict
Not sure this is the most appropriate error for the scenario, but hey, most of the interesting issues are never flagged appropriately in any microsoft software!
I´m using this workaround and it has worked for me:
Front end: Ms Access
Backend: Mysql
On the Before update event of a given field:
Private Sub tbl_comuna_id_comuna_BeforeUpdate(Cancel As Integer)
If Me.tbl_comuna_id_comuna.OldValue = Me.tbl_comuna_id_comuna.Value Then
Cancel = True
Undo
End If
End Sub
I just had very havy write-conflict problems (Acc2013 32bit, SQL Srv2017 expr) with a rather "heavy loaded" Split-Form.
For me - at last - was the solution to get rid of the write-conflict problems to simply
SET THE AcSplitFormDatasheet to READ-ONLY !!! (I haven't a clue why it was read-write anyway i must have set it by fault...)
It did nearly cost me a whole week to find that out.
I was having this problem and saving the record, marking Dirty to false, etc. did not work. It ended up that adding a timestamp column to the SQL table is what avoided/fixed the issue.
When last time I got this error, it was bit field having NULL value issue.
But this time, it was different text size of source table field vs linked table field.
I checked all my bit fields in various tables but didn't find any issue. All of them had default value, so there were no NULL values for bit fields. I observed that a text field with nvarchar(500) was giving this error. The linked table was using old field size 50 instead of recently changed 500. Relink of tables solved the problem.
So another finding is if the data type is changed for a linked table, you need to relink the table.
Just had this issue on MS Access 365 connected to PostgreSQL server. The error only occurred when trying to edit the first row.
I manually deleted the first row in pgAdmin 4, and then manually added it again. This solved the issue.
I was receiving the same error message.
Id Column in database table was set to BigInt, changing it to Int resolved the issue.
I ran a large query (~30mb) which inserts data in ~20 tables. Accidentally, I selected wrong database. There are only 2 tables with same name but with different columns. Now I want to make sure that no data is inserted in this database, I just don't know how.
If your table has a timestamp you can test for that.
Also sql-server keeps a log of all transactions.
See: https://web.archive.org/web/20080215075500/http://sqlserver2000.databases.aspfaq.com/how-do-i-recover-data-from-sql-server-s-log-files.html
This will show you how to examine the log to see if any inserts happened.
Best option go for Trigger
Use trigger to find the db name and
table name and all the history of
records manipulated
sometimes i face the following case in my database design,, i wanna to know what is the best practice to handle this case:::
for example i have a specific table and after a while ,, when the database in operation and some real data are already entered.. i need to add some required fields (that supposed not to accept null)..
what is the best practice in this situation..
make the field accept null as (some data already entered in the table ,, and scarify the important constraint )and try to force the user to enter this field through some validation in the code..
truncate all the entered data and reentered them again (tedious work)..
any other suggestions about this issue...
It depends on requirements. If the data to populate existing rows for the new column isn't available immediately then I would generally prefer to create a new table and just populate new rows when the data exists. If and when you have all the data for every row then put the new column into the original table.
If possible i would set a default value for the new column.
e.g. For Varchar
alter table table_name
add column_name varchar(10) not null
constraint column_name_default default ('Test')
After you have updated you could then drop the default
alter table table_name
drop constraint column_name_default
A lot will come down to your requirements.
It depends on your application, your database scheme, your entities.
The best way to go about it is to truncate the data and re - enter it again, but it need not be too tedious an item. Temporary tables and table variables could assist a great deal with this issue. A simple procedure comes to mind to go about it:
In SQL Server Management Studio, Right - click on the table you wish to modify and select Script Table As > CREATE To > New Query Editor Window.
Add a # in front of the table name in the CREATE statement.
Move all records into the temporary table, using something to the effect of:
INSERT INTO #temp SELECT * FROM original
Then run the script to keep all your records into the temporary table.
Truncate your original table, and make any changes necessary.
Right - click on the table and select Script Table As > INSERT To > Clipboard, paste it into your query editor window and modify it to read records from the temporary table, using INSERT .. SELECT.
That's it. Admittedly not quite straightforward, but a well - kept database is almost always worth a slight hassle.
I've got a new website moved to my server. This website uses PHP and MySQL, and is built very poorly.
The problem is - it needs non-null values inserted into new records where ever I don't specify a value, though the default value is null.
I've been told it has been done on the previous server, but they have no idea how. I'd be glad to receive some help on this.
You could update the default values of the fields of your database to prevent problems using:
ALTER TABLE `table` ALTER `field` SET DEFAULT 'value'
More information on ALTER TABLE for specific fields and parameters can be found in the documentation.
You need to add default values for the columns, either recreate the tables with defaults or alter the table definitions.
http://dev.mysql.com/doc/refman/5.1/en/alter-table.html
http://dev.mysql.com/doc/refman/5.1/en/create-table.html