|
actually, I'm thinking it might be a good idea to also set up all indexes (most importantly, the clustered one) on the blank new table before I do the bulk INSERT INTO.
this way SQL will be building the indexes along the way as it copies the data. I think this is potentially less time-and-space costly compared to SQL having to physically repartition the data in the clustered index later on, after the INSERT INTO has completed.
let's see 
|
|
|
|
|
udpate:
so even this way, the LOG file grows so much that it fills the whole 1000 GB disk. which is ridiculous as the db itself without the indexes is only about 110 GBs big. it's 315 GBs with all indexes. in the new table I'm only using 2 out of the 6 original indexes. as always when one does some change, I've found out I'm actually no longer using the remaining 4 indexes, so I dropped them in the new table
I think the problem here is that although the recovery model is set to simple, the log file keeps track of all the transactions until the statement has terminated and all data has been fully written to the disk.
so I think the solution now is to try to do this INSERT INTO in batches and make sure the log file gets truncated after the completion of each batch.
I'm wondering just how much disk space this operation would eventually need to complete if done at once. it would be useful if SQL server provided a way to carry out these bulk operations in some kind of "unsafe" mode with no logging at all. unless I'm missing something, it's not possible to turn off all logging in the MS SQL server.
I'll update you
|
|
|
|
|
final update:
all done. the correct way on how to do this was:
1) create a new table with the same columns, change Int32 column to Int64
2) script out all keys and indexes from the old table and create them on the new table
3) make sure the DB is in the simple recovery mode
4) INSERT the rows from the old table into the new table in batches, truncating the LOG after each batch. I was also inserting the identity column values, so I have also set the IDENTITY_INSERT to ON before the batch loop.
that's it
thanks for your help, guys.
Michal
|
|
|
|
|
Hi guys, I hope you can help me with this query. Maybe because it is saturday morning I can't seem to work this out. I have 2 tables, Sales and SalesDetails. To simplify things say the Sales table consists of ID, and InvoiceDate, and SalesDetails consists of ID, Quantity, SaleID, ProductID. Now I want to return a list of the total daily sales for a specific month for a specific product, i.e. 01/03/2010 - 5, 02/03/2010 - 10 etc. Hope someone can help
|
|
|
|
|
Try this
SELECT
SaleDate,
SUM(Amount)
FROM SalesDetail SD
INNER JOIN Sales S ON S.SaleID = SD.SaleID
WHERE SaleDate BETWEEN ThisDate AND ThatDate
GROUP BY SaleDate
Tricky bits will be if you have store time component in your SaleDate, you will need to strip the time off.
You can also use the month/year component of the SaleDate instead of the between but it is messier.Never underestimate the power of human stupidity
RAH
|
|
|
|
|
It worked. Thank you muchly. The dates are stored with no time component, so that was not an issue. The final SQL came out thus:
SELECT
s.InvDate, SUM(sd.Qty) as Amount
FROM SalesDetails sd
INNER JOIN Sales s ON s.ID = sd.SaleID
WHERE s.InvDate BETWEEN DATE(dteStart) AND DATE(dteEnd)
AND sd.ProdID = ProductID
GROUP BY s.InvDate;
|
|
|
|
|
I'm curious, what database are you using as this does not work on SQL Server
DATE(dteStart) Never underestimate the power of human stupidity
RAH
|
|
|
|
|
Hi
I am using MySql, querying it from VB.Net. deStart and dteEnd are VB variables. MySql does not like VB date variables, and so it is best to convert to date in your SQL Query.
|
|
|
|
|
Hi,
I have a few questions. First of all, what's the flexibility and strength of MSAccess 2007 and how much data and transactions can it support?
Second, I would like to know whether it's optimum to use MSAccess 2007 as database for a small banking system, provided it's a real life application and not a demo/tutorial one. The system will have approximately 250-300 customers per year and their daily transactions for the whole year that'll count upto 1,10,000.
If MSAccess isn't optimum then which light version DBMS can I use? How about MySQL or something else?
Third, is there any light version for MSSQL server that fits into my scenario, for almost all the MSSQL versions I have seen are developer/enterprise editions?
Regards
Priyamtheone
|
|
|
|
|
The strength of MSAccess is that it has a builtin GUI, a small footprint and xcopy deployment.
The weakness is that it doesn't scale, I wouldn't recommend it for more than ten concurrent users.
I would instead start with either Oracle Express edition or MS SQLServer Express edition. Both are for free, in both cases the biggest limitations are: won't use more than one processor, 1 GB Ram, and max DB size is 4GB.
This limits you less than you might think and both options are easy to upgrade."When did ignorance become a point of view" - Dilbert
|
|
|
|
|
I'm not sure about the 2007 version but Access was NOT a multiuser, it worked OK as a mutliuser DB until it didn't and when you asked for support from MS they would stand on the disclaimer that it is NOT multiuser.
Definitely use the Express versions (I use MSSQL but that is because of my skill set) and set up a proper client server environment. MSSQL is reasonably easy to set up and very easy to work with.Never underestimate the power of human stupidity
RAH
|
|
|
|
|
Imagine an app that where modification of values in the GUI are immediately updated to a SQL database.
Given this scenario, would you still open and dispose of a SqlConnection object for each update? My guess would be yes because the real connection to the database should be managed via connection pools.
Any rationale for either tactic would be greatly appreciated.
|
|
|
|
|
treefirmy wrote: Any rationale for either tactic would be greatly appreciated.
I'd open it just before the Connection is used, and close it as soon as I'm done with it; Take a look at tip #3 on Connection Pooling[^].I are Troll 
|
|
|
|
|
I have a database that holds information including isbn numbers for books. I discovered that the isbn is not specifically a 10 digit number it can actually have a character in it so i need to change it to an nvarchar but when i try to do so it says i can't that the table needs to be dropped. Surely I don't have to delete the table and reenter thousands or records just to change that property or do I?
Please help thanks.
|
|
|
|
|
if you can't change a field, perform these steps:
- add new field (with a new name)
- copy data from old to new field
- delete old field
- optional: rename new field
You can perform these interactively, using SQL Sever Management Studio, PHPmyAdmin, or whatever fits your database.
|
|
|
|
|
wow that worked perfectly. Luc i can't thank you and the other's enough for all your help you have given me. Hopefully one day I will be able to help others and return the favor in which you have given me.
Thanks again.
|
|
|
|
|
Someday saying thank you... that rated a 5!
|
|
|
|
|
Hi All,
I ran into an issue when trying to Decrypt a password field that I had previously encrypted and was hoping someone could help. I'm using SQL Anywhere 10.
I encrypted the password field with no issues, using the following trigger when a new record was added to the table:
ALTER TRIGGER "encrypt_new_user_pwd" BEFORE INSERT
ORDER 1 ON "QAS"."tableName"
REFERENCING NEW AS newPwd
FOR EACH ROW
BEGIN
Set newPwd.pwdField = ENCRYPT(newPwd.pwdField , 'key')
END
The problem is I'm unable to decrypt the password of the newly created record (fieldID = 0002), using the following:
SELECT CAST (DECRYPT(pwdField, 'key') AS VARCHAR(100))
FROM "QAS"."tableName" WHERE fieldID = '0002'
In Sybase when trying to execute the above statement I get the following error:
Interactive SQL
The following error occurred while fetching results:
Decryption error: Input must be a multiple of 16 bytes in length for AES
SQLCODE=-851, ODBC 3 State="08001"
Do you have any idea what's causing this error?
The pwdField is of type VARCHAR and size 15.
Thanks,
Mel
|
|
|
|
|
This is just a guess but if you read the error message it says:
MWRivera wrote: Input must be a multiple of 16 bytes in length for AES
But you say:
MWRivera wrote: The pwdField is of type VARCHAR and size 15.
Is 15 a multiple of 16?
Make pwdField bigger and see if that helps.
|
|
|
|
|
Hi David,
Thanks for the reply. I changed the length of the field to 16 and this stopped the error message from displying, the only problem then was the password was showing up as squares. I then changed the length of the field to 80 (5x16=80) sticking with the multiple of 16 rule and it works great now.
Thanks for your help with this it lead to me working it out. The reason I was recieving squares I asume is because the encrypted version of the password was larger than 16 and so the full encrypted version of it was too big to fit in the password field, therefore when I tried to decrypt it, it wasn't all there to decrypt and hence the nonsense squares.
Thanks again,
Mel
|
|
|
|
|
Hello friends,
which is the better way to use transaction?
From Database
or from Application?
|
|
|
|
|
There is no answer to this, it depends on many factors
I know the language. I've read a book. - _Madmatt
|
|
|
|
|
Hi,
I need some help with the database design for storing an expression tree. It should basically store a whole expression.
I'd want to query out data (in one or more result sets as necessary) so that I can finally evaluate it somehow in my front-end application. Just kind of need a table(s) structure to store and represent the relationship.
The expressions I plan to store in the database will be mostly logical expressions. Meaning they'd evaluate to a true or false mostly.
Here is a diagrammatic representation of what I want (as a result set(s)).modified on Thursday, March 4, 2010 8:31 AM
|
|
|
|
|
Hi,
I often use trees and what you essentially needs is the ID and Parent ID
for example:
ID | ParentID | Name
1 | 0 | A
2 | 1 | AA
3 | 1 | AB
4 | 0 | B
5 | 4 | BA
6 | 4 | BB
looks like:
[]
|
-A
--AA
--AB
-B
--BA
--BB
and so on...
... I hope you mean this 
|
|
|
|
|
I am using VB.Net 2008 and ADO.Net to do a Batch Update to our Oracle database.
The updates are working, but there is a trigger on the table before the row is updated to enforce a member's termination termination date.
So if I was trying to set the termination date (via the batch update) to 31-Jan-2010 but the member had a claim that was processed on 2-Feb-2010 the trigger would force the termination date to be 2-Feb-2010.
However, the trigger is NOT executing when the batch update runs?
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
