I have been asked to write a SQL script that can be run, which will stop replication and remove the subscriber/subscription.
Is this at all possible? Or do you have to use the GUI?
Which version of SQL Server you are using.
Try
sp_removedbreplication 'DB_PROD'
go
Related Read - How to cleanup Replication Bits - http://blogs.msdn.com/b/repltalk/archive/2010/11/17/how-to-cleanup-replication-bits.aspx
Google as well provides MSDN article in results
How to: Disable Publishing and Distribution (Replication Transact-SQL Programming) - http://msdn.microsoft.com/en-us/library/ms147921.aspx
If you're look to completely remove replication, including all the "bits". Or your server only has one publication and this is the one you're trying to remove.
I would recommend:
Connect to the Publisher or Distributor you want to disable in Microsoft SQL Server Management Studio, and then expand the server node.
Right-click the Replication folder, and then click Disable Publishing and Distribution.
Follow the steps in the Disable Publishing and Distribution Wizard, and instead of processing select the option to generate scripts.
This has the advantage of not only completely cleaning things up. But also removing the distribution database, which if you've ever dealt with before is known to become littered with remnants of replications past.
Full documentation can be found here.
Related
On my development box, I always run the latest version of SQL Server. I often deploy databases from my dev box to a live/staging area for review or testing. I've done this many times and it has always been a painful process, but I am realizing that I need to find an easier, more reliable and consistent way of performing this basic operation.
I normally use WebMatrix purely for deployment and it's worked fine, but I've been having problems getting it to work on my server for some reason. Consequently, I am seeking an alternative solution.
Creating a SQL dump file would probably work, but it's not an acceptable solution a database contains images and easily exceeds 2 gigs of data which would take forever.
The Import/Export utility fails due to issues with incomplete schema copies, identity inserts and checks. The solutions offered for these issues has failed to work in my particular case.
The Backup and Restore method also fails due to some strange incompatibilities between SQL Server 2008 and 2012. SQL Server 2008 Management Studio throws exceptions during the restore process of a 2012 database. It's odd that this happens, even though I set the compatibility of the database to version 2008.
I haven't tried the Detaching, copying and reattaching files method, but I haven't bothered trying since it would probably fail for the same reasons the backup and restore method did.
Are there other alternatives out there? Also, why is this so unbelievable hard for a task that is so common and important, especially in this day in age of 2013? Get real Microsoft!
We changed our method of deploying and moving databases between servers, instances and versions by adopting the tools from RedGate. They are expensive, but worth it IMHO.
My team creates scripts for ~everything.
Database Creation, Alter, Inserts, etc, etc.
And we write all scripts that check for the existence of things before trying to create them.
Aka, we can run the scripts over and over and get the same results.
And we deploy to different environments by using SqlCmd.exe.
EDIT
See:
http://odetocode.com/blogs/scott/archive/2008/02/02/versioning-databases-views-stored-procedures-and-the-like.aspx
and
http://odetocode.com/blogs/scott/archive/2008/01/30/three-rules-for-database-work.aspx
=============
If that is "too much" then I agree with the other poster, RedGate is your friend.
Points below aside, have you considered the Database Projects within VS2012? they allow you to script off the tables, sp's, triggers, users etc you want, generate sql cmd scripts, make changes and schema compare and version control your database code, I'd certainly recommend it
"Creating a SQL dump file would probably work, but it's not an
acceptable solution a database contains images and easily exceeds 2
gigs of data which would take forever."
Why is this a problem? where are you transferring the file from and to and over what connection?
"The Backup and Restore method also fails due to some strange
incompatibilities between SQL 2008 and 2012. SQL 2008 Management
Studio throws exceptions during the restore process of a 2012
database. It's odd that this happens, even though I set the
compatibility of the database to version 2008"
This shouldn't be an issue if file is created in 2008 prior to restoring. If you create a new DB in your 2008 instance, then take a backup from that and restore it to a 2012 instance with 2008 compatiblity, then you should be able to use it there, back it up from the 2012 instance and restore to 2008 again afterwards.
I am new to replication. I have two sql server 2008 servers running on windows 2008 R2. The servers are in two different locations and on two different domains. I have been able to use aliases to get both "Snapshot publication" and "Transactional Publication" working perfectly. But what I need is "Transactional publication with updatable subscriptions" so if a change is made on either server, the changes are replicated to the other server.
When I run through the New Publication, I get through every page to the very end with no problem but when I click the finish button I get an error. There are three actions and it fails on the first action called "Creating Publication 'xxxx'" The message I get is "SQL Server could not create publication 'xxxx'. an exception occurred while executing a Transact-SQL statement or batch. The distributor has not been installed correctly."
I have searched for an answer and cannot find it. I think this is a permission problem between the two servers but I have no idea how to solve it.
Any help would be appreciated.
In my experience the installation of different types of replication over the top of one that was previously implemented can cause issues.
If able I would suggest clearing all replication and starting from scratch with your new approach.
You have to run a variety of stored procedures to get it completely off the server. Using the GUI only doesn't do nearly as good a job of cleaning everything off.
This guide from Microsoft should get you started.
http://msdn.microsoft.com/en-us/library/ms152757.aspx
Hi My company is deciding for switching its existing application to azure platform (only Sql Part). So we need to upload our db from local to cloud. For migration i came across various tools like
1. cerebrata 's tools
2. SqlAzure Migration wizard
3. Microsoft Sql Data Sync
4. Conventional Script way via management studio.
But all the above tools showed that they have limited capacity. A user cannot work flawlessly on either of the tool.
In cerebrata's tool - the main drawback was its field for Application User Name and Application Key , which my admin havent shared. Also there is manual mapping of fields between azure and local.
Sql Azure Migration wizard - generates scripts and executed too but with lots of error . I was using its version 2.1. Also it very slow. It seems that its a replica of Sql Srvr Mgmt Studio.
Sql Data Sync :- I found it cool as its a MS product but it has limitation too that it only connects with Windows Authentication based local sql server, or you need to explicitly allow the required but. Even after allowing while syncing , I got some Sql Azure Provisioning Error.
4 Sql Srvr Mgmt Studio :- This is most easiest way but requires a lot of manual work to do before actual migration. What i did is that I generated a script of entire db (almost 101123 lines of code for single db) and tried to execute on azure. On the very first time i faced some keyword mismatch error . Finally i removed all line after primary key declaration that With (Padding = Off ....)or something similar and also On Primary then i executed , but still got error on Set Identity Insert On. After doing a lot of hard work in removing unwanted lines waited more than 2 hrs to completed the script remotely, i got no Errors , errors and errors.
So you guys are requested to please suggest me any good alternative stated than above or i am lacking something and can do more with above.
Thanks
Amit Ranjan
I've faced a similar problem recently, running through the options you've listed.
You might give a try to Red-Gate beta for Azure (free for a few months). I found their tools to be quite good for SQL schema and data replication.
Never tried the Azure build myself, though (I migrated tables manually by the time I was told about the offer).
Has anyone come across any good tools (preferably but not necessarily FOSS) that can read a SQL Server (2005/2008) trace file and execute the commands against another database. We are attempting to perform some performance testing on our SQL servers and would like to replicate an actual load.
I have come across but not yet used:
JMeter
ReplayML
Preferably, the application would be able to use threading to mimic user connections and query execution on the SQL Server.
You can replay a SQL Server Profiler trace against another server using the SQL Server Profiler itself.
See the following Microsoft Reference as a starting point.
http://msdn.microsoft.com/en-us/library/ms189604.aspx
Quest Software also have a tool called Benchmark Factory that can be used to perform SQL Server load testing.
http://www.quest.com/benchmark-factory/
One of the best tools is actually freely available from Microsoft. The RML Utilities are targeted at SQL2005 & SQL2008 and are specifically designed for this type of testing.
You can download the tools from http://www.microsoft.com/downloads/details.aspx?FamilyId=7EDFA95A-A32F-440F-A3A8-5160C8DBE926&displaylang=en
We have used them to solve several performance and locking issues.
Note: Capturing trace files using the SQL Profiler GUI can add to performance problems due to the way the GUI and Trace backend interact. The RML Utilites include a script that can capture traces directly from the SQL Server without using the GUI.
You can replay trace files directly in SQL Profiler, although I've only used it a couple of times for that, so I don't know what all of the limitations are on it.
team system has an add-on that you can find on codeplex: It is called ->
sql load test
Let me know if that works well for you.
I know this is a really old question but after searching for some time i discovered a new open source tool.
https://github.com/spaghettidba/WorkloadTools which works great
I have a live database on a shared hosting server. I am making some major changes to my site's code and I would like to fix some stupid mistakes I made in initially designing the database. These changes involve altering the size of a large number of fields, and enforcing referential integrity between tables properly. I would like to make the changes on both my local test server and the remote server if possible.
I should note that while I'm fairly comfortable with writing complex queries to handle data, I have very little experience modifying database structure without a graphical interface.
I can access the remote database in the visual studio database explorer but I can not use that for anything other than data manipulation. I installed Sql Management Studio express last night and after 40+ crashes I gave up - I couldn't even patch the damn thing.
The remote server is SQL 2005 / The MyLittleAdmin web interface is available.
So my question is what is the best way to accomplish these changes. Is there a graphical interface I can use on the remote server? If not is there an easy way to copy the database to my local machine, fix it, and re upload? Finally if none of the above are viable does anyone have links to a decent info on fixing referential integrity via query?
Sorry for the somewhat general question - I feel like I am making this far harder than it should be but after searching / trying all night i haven't gotten anywhere. Thanks in advance for the help. I really appreciate it.
...Also does anyone have a time machine I can borrow- I need to go kick my past self's ass for this.
Usually hosting providers allow you to backup and restore your database, so the easiest way to accomplish the move is to backup your live DB, download the backup file, restore it locally, do all the changes, do a backup of the local db, upload it, then restore it in the live service. Your site should be placed on an administrative shutdown during this time so it does not continue to update the data while you're doing this operations. You have to make sure your local SQL instance is exactly at the same build version (##version) like the hosting provider, otherwise your local SQL may upgrade the database structure and you'll be unable to restore it back on the hosting provider (or you'll be unable to restore in on your local server if your version is earlier than the host's). The MSDN BOL has a detailed guid on how to Copy Databases using Backup/Restore.
An alternative to backup/restore is to detach/attach the database, but I do not recommend this because you need to move both the MDF and the LDF in sync, and they're also larger in size than a backup.
This assumes you can do all the schema changes on your local copy in a wizardly manner, ie. fast and correct. Of course, that is not easy. The recommended way is to prepare in time a script that applies all the transformations needed to reach the new schema. There are tools like SQL Diff, SQL Compare, SQL Delta and other that can generate such a script. Also Visual Studio Database Edition can do this.
How I would do this would be like this:
Ensure I have exactly the same schema on my dev machine as on the live host. If not sure, I can take a backup of the live server and restore it localy. This would be my reference, v1. schema.
Keep the backup of v1. for reference
Start developing a script that changes the schema to my target. Sometimes I need to refresh my memory on script syntax myself, and what I do is I go to the SQL Server Management Studio wizards for the operation I want to do, select all the options in the UI and then select the 'show script options', that will show me exactly the script SSMS is running to accomplish my desired change.
For each change I add to the script, I can test it by restoring the v1. reference backup I have from step 1 and running the script.
Keep iterating on the script, adding one change at a time, until all the needed schema changes are done. After each change, I can test it again like in step 4.
Yourscript should do not only DDL changes to the schema, but also any DML changes needed (modifying reference data, changing values, moving columns between tabels etc).
When the script is ready, I can download a newer backup, apply the script, and upload the updated backup and restore it on the live host. Alternatively you can simply run the script on the live host (after of course you backed it up in case something goes horribly wrong).
In my projects I always rely on scripts to deploy and upgrade the database. In fact I use the database extended properties to store a 'version' of my application deployed schema and in my code I simply roll forward all the scripts that bring the schema to my last version. I have an article on my blog describing this technique: Version Control and your Database.