I have a creation script for a SQL Server database and beside it I have a test script
My test Script should compare the result of my original script and the database and let me know if it is a match
I need to store the resulting script in a repository.
I've already got a script and the database. It's not hard to write the test script, but I want an app that automates the process.
I've already got Redgate tools and I heard it can make the job for me, but how? If not, are there any other tools that could do such a thing?
If you're looking to compare the structure of the two databases this can be done using Redgate SQL Compare.
Set the location of your script as a "Scripts Folder" source within the tool.
Set the database where you have run the script as the target.
Run the comparison. This will show you if there are any differences.
Generate a report of the differences which can be referred back to at a later date. (https://documentation.red-gate.com/display/SC12/Generating+a+report)
To automate this process you can make use of the SQL Compare command line. (https://documentation.red-gate.com/display/SC12/Simple+examples+using+the+command+line)
Full disclosure: I'm a developer at Redgate for SQL Compare.
I'm not sure if you are searching something to compare your data or the structure of your databases.
If it's for the structure (like Red-Gate SQL Compare and not SQL Data Compare), you can install SSDT (Microsoft SQL Server Data tools - It's free and available for visual studio).
With SSDT you will be able to create a database project and compare 2 databases structures.
Here is a quick presentation that I found on youtube:
https://www.youtube.com/watch?v=FKSsbWnFMD8
You can completely automate comparison and reporting with Apex Diff. Not sure which SQL Server are you using, but it's free for SQL Server Express.
Here is a step by step tutorial, which you should follow with one difference - set Script folder as data source:
https://solutioncenter.apexsql.com/how-to-keep-a-source-control-repository-updated-with-the-latest-sql-server-database-changes/
Related
I have an existing database (SQL Server 2008 R2), and my goal is to have a project that can install the same database as I already have.
i.e the project will read the schema and the data, and will be able to duplicate the existing data and schema.
This tool needs to save the database information to disk so I will able to install the database on different machines.
Can I do it with Visual studio 2010 database projects?
Is it a simple task?
(The reason I am asking is that I already have such a project that installs my database, but this tool is very old(~2000).
That old project is one big mess, it has C#, java, sql, batch files and some 3rd party tools and has many more capabilities than I need for my current product..)
Thanks.
Use SQL Server Management Objects (SMO) to create a snapshot of your source db and then restore it to the target server.
More on SMO here: http://msdn.microsoft.com/en-us/library/ms162169.aspx .
A nice tutorial here: http://www.codeproject.com/KB/database/SMO_Tutorial_1.aspx
The easiest way is to backup using SQL server management studio. Then move the backup file to the second server and there restore it with SQL server management studio. If you need SQL code to do it (instead of GUI way), just drop a comment...
This codeplex project has a basic library that does that if you like to roll your own and has two sample projects for migration/comparison tasks.
http://dbschemareader.codeplex.com/
Also the commercial option is to use Sql Redgate's tools. They are expensive but worth every penny if this is a mission critical database. On the otherhand, if you are just using a db as an alternative to a file, say in a embedded scenario, it might be hard to justify such an expensive tool.
Have you tried DBSourceTools ? http://dbsourcetools.codeplex.com
It is an open-source database versioning tool that will script out all database objects and data to disk, and then re-create the database on another machine.
It has all of the features that you are looking for.
Is there a tool that will let me generate a single script containing all tables and views? Sql Publishing Wizard drops everything (so all data is lost) and recreates it. It does have an option to not drop, but in that case, it doesn't update tables that exist (if any columns have changed).
In SQL Server Management Studio 2008 you can right-click on a database in the object explorer, go to Tasks > Generate Scripts..., and that will give you the option to choose not only what object types you want to script, but whether or not you want to script the drop as well.
When you are making changes to existing tables, you should be writing alter table scripts to make the change and then putting them in source control like any other code. Then when you deploy a set of changes, you run the scripts you created for that deployment.
Otherwise, yes use SQL compare.
Should you look at the Red Gate products specifically SQL Compare. They'll handle any situation you could need concerning script generation and database synchronization. (You can get a trial license too, to try it out and see if it is what you need.)
Have a look at this tool can be used which has the capability to generate the create and drop scripts for the SQL server objects, provided in a configuration file.
This tool uses the same mechanism as SSMS tool uses to generate the script.
SQL Server Script Generator Tool (via C#)
Have a look at these:
WinSQL (Lite edition is free, other versions are reasonably priced + free trial)
OpenDbDiff (free)
You can also check out MyDbUtils which can create scripts for:
Stored Procedures
Functions
Views
Triggers
I have a 2 MS SQL 2005 databases,a TEST and DEV database. Now our developer added some extra columns,tables etc in the DEV database.This created differences in the TEST database.is there a script i can write tha can tell me what the changes where in the DEV database between certain dates...i found a couple of tools but they are quite basic and dont really generate change scripts etc. Also tried the change script function in management studio but it seems to be working when the change is first made and not later.
Appreciate your thoughts.
A.
redgate SQL Compare is a good tool to check for differences between databases and even sync them.
However, if you need to do this for free, try this: within in SQL Server Management Studio do the following:
1) script the entire schema of TEST to a file, look here if you don't know how
2) repeat step #1 but for the DEV database
3) diff the files using something like Beyond Compare, should have a 30 trial
I wrote dbscript, and one of its features is to compare two database schemas and create a migration script.
What you need is a way to manage changes to your database schema and then apply them in a controlled and consistent manner. Moreover, you need a single authoritative source for the database schema.
For all this, check out Wizardby:
(source: googlecode.com)
Is it possible to restore a backup of a SQL Server 2008 database onto an instance of SQL Server 2005?
I need to work on an sample application for which database backup is in sql server 2008.
But I'll not be able to install 2008. So is it possible to restore that back up in 2005?
No. It is not possible to restore a database from a backup of a newer version.
If you are dead set on it, I think your best option is to selet the database in the Object Explorer in SQL 2008,
right-click, select Tasks->Generate Scripts.In the options dialog emable about everything, including Script Data.
And make sure you select "Script for SQL 2005".
Source
When importing the objects into your target server, if the objects are large you may find that you can't open the SQL file via Management Studio (with a completely useless "The operation could not be completed" error, no less). That's okay, just load the file via sqlcmd.
One important thing is missing in all answers and that is the fact that Generate Scripts in SSMS doesn’t order the scripts correctly.
Scripts have to be ordered in the correct dependency order so that child tables are created after parent tables and such.
This is not an issue for small databases where its easy to reorder the scripts manually but it can be a huge issue when dealing with databases that have 100+ objects.
My experience is that its most convenient to use third party tools that can read backup and generate scripts in the correct order. I’m using ApexSQL Diff and Data Diff from ApexSQL but you can’t go wrong with any popular vendor.
No, not directly. SQL Server 2008 database backups are not backward compatible with SQL Server 2005. However, with SQL Server 2008 Management Studio, you can script data and schemas in SQL Server 2005 mode. This article describes the process in detail.
Yes it is possible
Using the export in the SQL Server 2008. Go to All Programs --> Microsoft SQL Server 2008 --> Import and Export Data
Then SQL Server Import AND Export Wizard window will be opened. Press Next
Choose a Data source (in your case from SQL Server 2008). Choose a Destination( in your case from SQL Server 2005).
Select Copy data from one or more tables or view
Select the source's tables and destination's tables
Click Next & Finish
to complete.
I have had this problem for a long while.
You cannot restore SQL2008 backups onto an SQL2005 instance.
And for me, workarounds like import/export wizzard or to script the database from SQL2008 using the generate scripts with the for SQL2005 option won't work.
Because simply, my databases cross-reference each other inside their views and stored procedures or udfs. They do not befall to my responsibility completely and so I cannot consolidate them into 1 database.
They are a set of 6 dbs that refer to each other directly inside their views and stored procedures.
When I transfer them from one SQL2005 instance onto another, I usually do full-backup/restore.
If I were to script them, even with the with dependencies option I would get errors at re-creation time as db1 will not find views inside db3 because it so happened that I executed the create db1 script first. If I tried db3 first I get similar exceptions.
The only way to script them so that I won't have such dependency exceptions, is to figure out a sequence that works and script them partially in that manner: say: db1_tables followed by db2_tables followed by db2_views followed by db1_views, sp, udfs etc.
I have such a sequence. And when I need to create a new set of such 6 dbs, I executed the smaller partial scripts in sequence.
This explains why the generate scripts, with dependencies and with data and set to SQL2005 version scripts, will just not work for me.
The import/export wizzard is better in my case because it will copy tables and then you still have to script all views, sp, udfs etc.
What I really need is a conversion tool for SQL2008 backup files, to convert them to SQL2005 format. Then my problem will go away.
Or some kind of a tool that would allow restore from SQL2008 full-backup files, without asking me too many questions.
If anyone knows such tools and have used them, let me know.
You can use DBSave, it's a great freeware tool to backup and restore ms sql server on different machines.
It's verry simple to setup and to use.
No you can't, but tools like red gate's SQL Compare/Data Compare can read backup files directly & transfer the info across to a live database, dealing with any syntax or settings that aren't compatible on SS2005
Having had no luck with the Import/Export stuff (flat file exports failed on import claiming charset mapping issues [even though same charset used throughout] and/or truncation issues [even though source and destination had exact same structure]), and having had no luck with using the "generate scripts" option suggested by Garry Shutler (it generated a script with syntax errors), I was finally able to copy the big table I wanted to copy from 2008 to 2005 using the SQL Server bcp utility. So that's another option for this situation, although for an entire database it would be table-by-table and probably doesn't help with views and such.
The steps I used:
On the source server, use "Script Table As...CREATE" to get the structure, run that on the target server.
On the target server, create a bcp format file using your newly-created table:
bcp database.owner.table format nul -f table.fmt -n
(If you're not using Windows auth, you may need the -U and -P options to specify username and password.)
Copy that format file to the source server (if necessary).
Export the data to file on the source server:
bcp database.owner.table out table.dat -f table.fmt
(Again, possibly with -U and -P.)
Copy the data file to the target server (if necessary).
Import the data on the target server:
bcp database.owner.table in table.dat -f table.fmt
(Again, possibly with -U and -P.)
In hopes that proves useful to someone else.
I have been developing in VB.NET and SQL Server 2008 for a while now, but haven't got into live installs yet. In the database system I used be on it had the ability to archive multiple tables into a .dga file, as it was called. I could then restore the .dga file into another database or on another server.
I'm looking for the easiest way to accomplish something similar in SQL Server.
If you want to transfer specific tables, then using Data Transformation Services (right click on the database in SQL Server Management studio and select "Import Data" and it will bring the dialog up for it). Of course, this assumes that you have both databases available to you.
If you are comfortable with replacing the database as a whole, you can easily backup the database and then restore it into a new one through SQL Server Management studio (or through calling the appropriate SP).
I would go for one of the following :
From MS SQL Management Studio, right click on the database / Tasks / Generate scripts
From Visual Studio, in the Server Explorer tab, "publish to provider"
Both will launch a wizard allowing you to export the tables you want the way you want (including data or not, creation scripts or not, etc etc.)
If you want to move tabless without data, the simpliest thing is to script the tables you want and run the script.
We script all our db changes and commit them to subversion and then run them as part of the deplyment process.
If you want to put the whole database on prod including data (scrub out test records first!), then do a backup and restore onthe other server.
For future changes, wescript all our db changes and commit them to subversion and then run them as part of the deployment process. There also are tools that look at the structural differnces bewteen the two servers and creates scripts. REd-Gate's SQL Compare is really good for this.
In addition to HLGEM's suggestions, you can look into SSIS if this is an ongoing process.