|
Hi All,
I have a stored procedure that is inserting some values in Master and details table and I am getting the following error. Some times I am getting this error but some times I am not getting this error. When I am inserting detail table values I am checking everything if I am inserting any values that are not there in the Master table, but I am not getting any sign that I am inserting any values that are not there in Master table.
The INSERT statement conflicted with the FOREIGN KEY constraint "FK_Identifier_Entity". The conflict occurred in database "IdentityResolutionService", table "dbo.Entity", column 'EntityId'.
My scenario is: I have Entity table, Identifier table, Entity is the master table Identifier is the detail table. I am actually taking data from another table called Students to load all the students of a state, sid, dsid and ssn are the columns that are important from the students table. One sid may have multiple dsids depending upon year, school district etc, one ssn one sid. Now I have taken Rank on the SID to load all this data from Students table to Staging table. Now I have taken all the distinct SIDs and put it into Master table.
In the Identifier table when I am loading the SIDs, DSIDs and SSNs, I have taken the corresponding Rank value for that row, instead of creating one. I explained this only because there is no way that I can miss the Foreign key relationship. But one thing when I am loading the Entity values I am making it as Auto increment off, insert statement and then making it on because there is another application that is using the same table as auto increment on. But another thing is as the user doesn't have alter table permissions, I have created a Stored Proc which takes table name and insert statement as parameters so that I can make auto increment off and run the insert command in the parameter and makes auto increment on. After doing this I make auto increment on. This stored procedure runs under proxy user account.
Then I run insert into Identifier table, some how the insert into Identifier table is failing with the above error message. What could be the reason. Some time it works, some time doesnt. Is it because of dynamic sql or proxy user or the huge amount of data that is being processed because of all the students in a state, like around 4 or 5 million records am processing. In the Identifier table it is having more than 10 million records or am I doing any mistake. Because of its hugeness I am unable to debug the Stored Proc also very well the reason is it runs for an hour or something. All the time is going just in testing.
Is there any other reasons apart from unavailable values in master-detail tables for Foreign key constraint failure.
Please understand me and help me whatever you can, that would really great.
Thanks & Regards,
Abdul Aleem Mohammad
St Louis MO - USA
|
|
|
|
|
indian143 wrote: I am loading the Entity values I am making it as Auto increment off,
I suspect this is the cause of the problem, I can't see any reason for removing the IDENTITY attribute and the error message points directly to this as issue.
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
Can I put my code here
Thanks & Regards,
Abdul Aleem Mohammad
St Louis MO - USA
|
|
|
|
|
indian143 wrote: I explained this only because there is no way that I can miss the Foreign key relationship
But the error is telling you SPECIFICALLY that that assumption is wrong.
Hypothetically how do you know that, for instance, the SSN was never entered wrong? Or that someone didn't make up a SSN?
The error is telling you that either you must clean the existing data before using it or you must use another key.
|
|
|
|
|
I am not using SSN directly as foreign key what I am using is the rank generated depending upon the SID is foreign key means for each SID there will be an EntityId and the same Entity Id I will be using for loading the DSID and SSN too. There is no way that I will go beyond Entities that are created by SID's Rank.
I can give you the code, can you please look into it.
Thanks & Regards,
Abdul Aleem Mohammad
St Louis MO - USA
|
|
|
|
|
hi to all
whats diffrence between two below query in performance respect:
query 1:
SELECT n, dbo.AddOne(n) AS r
FROM dbo.T1
ORDER BY r;
function addOne is this:
CREATE FUNCTION dbo.AddOneInline(@n AS BIGINT) RETURNS TABLE
AS
RETURN SELECT @n + 1 AS val;
GO
and query2:
SELECT n, (SELECT val FROM dbo.AddOneInline(n)) AS r
FROM dbo.T1
ORDER BY r;
and defination of function is:
GO
CREATE FUNCTION dbo.AddOneInline(@n AS BIGINT) RETURNS TABLE
AS
RETURN SELECT @n + 1 AS val;
GO
query 2 is very good in performance and i want to know whats difference between in this two query
|
|
|
|
|
The difference is that one is written out as a subquery.
It's a micro-optimization IMHO, but you can use the Management Studio[^] to check the execution plan and compare both versions.
Bastard Programmer from Hell 
If you can't read my code, try converting it here[^]
|
|
|
|
|
Hi everyone,
I wanted to ask a question that was confusing me how to design the db.This bridge database that i was going to design has 3 different districts(Locations) say A,B and C.These districts have different road segments,sections.I temporarily designed the db using separate tables but having the same fields.As i said these tables only differ by the thing i mentioned.
I know I'm not following the design rules,so i need you to help me on that.
Thanks
|
|
|
|
|
And when the next district is added, you add another table?
No! Create a table for the districts, and then use "foreign keys" to connect to the sections table.
|
|
|
|
|
Bernhard Hiller wrote: foreign keys
What be these foreign keys you speak of!
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
BTW there are fields that provide the bridge information,it's bridge number(p.k),name,design year,asset value and other fields.
Now there are 3 different districts A,B and C.Under District A,there are different sections and road segments that correspond to A only.Same thing goes to B and C.These districts are not related in any way in other words.
That's why i created 3 separate tables based on these districts.
|
|
|
|
|
Nebilo wrote: I know I'm not following the design rules,so i need you to help me on that.
How? Yes, I could sum up some widely used constructs, but that's not a decent alternative for a good design.
There's explanation on database-normalization in the wikipedia, and various articles on CodeProject. Alternatively, you could fake a design[^].
I'll gladly help with the normalization rules if you get stuck 
Bastard Programmer from Hell
If you can't read my code, try converting it here[^]
|
|
|
|
|
hi to all
when i run this query i faced with timeout
DELETE FROM bml.LoanInstallment WHERE ID IN (select * from wfInstanceNodes where ID=@nodeID)
this occurred when nested select return null and i think it is because of that i have no index on id;
and i set index to id but still this query faced with time out
note that bml.LoanInstallment has 11000000 record and this query run without problem for a table with
a few records.
thanks for any help
|
|
|
|
|
Do you run into a timeout when you do a SELECT instead of a DELETE? E.g. SELECT COUNT(ID) FROM bml... etc.? And how many entries are to be deleted?
Are there indices for the ID columns in both tables?
Are there any constraints, e.g. an OnDeleteCascade?
|
|
|
|
|
id is uniqueness and
and one entries finally will be deleted.
and i don't have any constraint on bml table
thanks in advanced
|
|
|
|
|
Your subquery should not select "all" columns, only the ID-column.
Bastard Programmer from Hell
If you can't read my code, try converting it here[^]
|
|
|
|
|
ok
i use select id instead of select * but im still face with timeout
i think that this is because of that my nested select return an empty column
what you thinks ?
how can i handle this problem?
thanks for answer
|
|
|
|
|
mhd.sbt wrote: i use select id instead of select * but im still face with timeout
A command-timeout or a connection-timeout? How long does the query take when run on the server directly?
Can you post the schema of the tables here? It would be weird if you'd select the same key that you're filtering on.
mhd.sbt wrote: what you thinks ?
I'm pretty sure that you can select "where in null" without hitting performance-problems of said magnitude.
Bastard Programmer from Hell
If you can't read my code, try converting it here[^]
|
|
|
|
|
thanks for your time to answer this problem.
i found problem
that i posted it in a solution
|
|
|
|
|
Personally, I don't like sub-selects (they seem so 90s), I'd rather JOIN them.
(Provided your system allows it)
-- Transact-SQL extension
USE AdventureWorks2012;
GO
DELETE FROM Sales.SalesPersonQuotaHistory
FROM Sales.SalesPersonQuotaHistory AS spqh
INNER JOIN Sales.SalesPerson AS sp
ON spqh.BusinessEntityID = sp.BusinessEntityID
WHERE sp.SalesYTD > 2500000.00;
|
|
|
|
|
I have installed SQLServer2008R2 on my Windows 7
PC successfully.
However, when I open the configuration manager,
the "SQL Server Services" entry shows an error message saying
"The remote procedure call failed 800706be".
|
|
|
|
|
|
Hi all
In c# forum i asked the below question
Hi all
Please clear my doubt.
I have one server system where i installed sql server.
5 client machines accessing that.
now suppose if there is any power fluctuation like it's coming for 5 min and again going and so on.
So for this kind of situation, how to save the data or how to update the database each second.??
Is it possible that at a time we update the client database also and server database also.?
i.e at button click it should update both client and server database..
Please help..
Then i got the suggestion of database mirroring / database replication.
I am very much new to this.
Please need suggestion from you guys.
|
|
|
|
|
Code-Hunt wrote: Please need suggestion from you guys.
Google the topic. And buy a book or perhaps two.
Then set up some test servers. And test it. After you figure out the pro/cons of the possible different ways to do this.
Naturally of course would also need to consider other types of failures, which means that you need the two server located in different geographic areas. That of course means you must not only consider the methodology of replication but the cost. And find hosting companies to host at least on of the servers.
You must also test it and insure that the servers continue to run in production.
With the cost you might also consider the possibility that doing all of this yourself might not only cost a lot (if you are being paid) but will also have a high risk. Consequently finding a hosting service that provides all of the above but be more effective.
|
|
|
|
|
Get a UPS[^] for all affected systems including switches.
Be excellent to each other. And... PARTY ON, DUDES!
Abraham Lincoln
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
