Versioning - Deploy Just The Changes To The Server
Jan 8, 2008
At ScottGu's Blog about "Database Publishing Wizard", AlexD from codeplex said:
"Regarding the multiple requests for versioning, backup/restore of remote database, and selection of individual objects - these are all things we are actively looking at for our next release in the first half of 2007."
After so many searches, I still don't know if this tool performs Versioning, i.e, when deploying the database, just update de diferences between the local and server database.
Did Visual Studio Express 2008 have somethingh like that? (I know that VS Team Edition 2005 had).
If this tool can't make it (versioning), which tool/method did you recomend me?
Am getting errors trying to deploy a dtsx created by ms (the reporting services execution log one) to which I have made zero changes, but it is not working (2 errors shown below)
error from deployment wizard: ===================================
Could not save the package "H:SSISRSlogRSExecutionLog_UpdateinDeploymentRSExecutionLog_Update.dtsx" to SQL Server "xxxxxxxxxxx". (Package Installation Wizard)
===================================
The SaveToSQLServer method has encountered OLE DB error code 0x80004005 (Login timeout expired). The SQL statement that was issued has failed.
Dear friends,In the area of GIS (Geographic Information Systems) there is a featureknown as versioning (long transactions).This feature allows databases to maintain different versions of data,in a hierachycal structure, in order to do simulation (what if ...),historical snapshots, concurrent editing, etc.Each version can be reconciled with its parent version in any moment(merge-post changes).I have recently seen that Oracle supports this feature from version9i.I am very interested in knowing if SQL server will support thisfeature in future versions. Looking at SQL Server 2005 doc. I haven'tseen any related info.Thanks,Jerry
I'm having problems to keep proper versioning control in place between the development and production environments. I'm running SQL2000 and we have hundreds of packages that runs daily. Some on shedules and some not.
Every time a package is saved, it creates a version in SQL Server. After development I want to be able to use something like "Visual Source Safe" and book the final version in that was moved into production. Something with a version number etc.
This is especially a problem if I want to roll back to a prior version of a package. I do not know which one of the 1000 versions to choose from that were created while developing the package.
Another problem is that I do not know if someone is working on a package if I want to work on it.
I can not run a search on all the packages to get a list of which tables/fields are used where to determine the impact of a program/database/design change that needs to be implemented.
I have a big delimma. We are developing and application that requires parallel work to be done one different copies of databases at the same time. Then when one when group is done and ready to ship their bug fixes/features, the changes they make to the database and data have to be merge back into the baseline database.
Here's the specifics. 4 databases (that make up the product), 4 copies of the 4 database (one for each team).
I was thinking about using SQL DMO to attach to each database and comare each table's schema and data against the baseline (the current release) then scripting out the changes that were made.
Can someone give me some tips on how to maintain parallel database development and the merge process that can make this happen?
I replicate (transactional replication) my data entry database to aread-only database. Both are SQL 2000+SP4. The web server reads theread-only database. At times, there will be lots of changes in the dataentry database, thus lots of replications to the read-only database. Iam concerned that the replication may lock the data in the read-onlydatabase, causing slow response to the web server.I would like to use row versioning so that the read-only database cansupply old data when the same row is being written by replication. Iread that row versioning is a feature in SQL 2005. Is there anyversioning capability in SQL 2000?Thanks
I have a question about how I should go about handling different database versions (schema changes) with my application. I am using an sdf database as a local data store (either on the .NET framework or Compact Framework).
I set it up so that the database file has the database schema, but no actual data, which is copied to the AppData folder if it isn't already there. The I load the database into the dataset, and can store data in the database with no problems.
What I want to figure out is what happens when I later decide to change the database schema. For example, say I add a column to a table. When I load the existing database into the dataset, I get an exception because the existing database doesn't have that column.
It seems that there should be some way to update the existing database so that it adds the column into the table, and sets the rows to just have NULL for that new column.
I am not sure if the TableAdapter or some other object should handle updating the existing database so it matches the latest dataset schema, or if I need to manually write SQL statements to modify the existing database.
I was wondering if anyone knows if there is a way to version a report, after changes have been made kind of like there is to an application. when changes are made to an app with a version # of say 4.0.1 you can changge to 4.0.1. Without using a program like SourceSafe. Thanks in advance for the speed responses.
Could someone point me in the right direction? I have an internal development database and a production database. Is there an easy way to replicate the changes that have been made to the development version on the production server without modifying the actual data in the tables? So, if I add a new user in my development version I offcourse don't want to see it pop up in the live version. But adding/deleting/updating a new table or column should.
And if possible I'd also like to know how you could do the following: Let's say we have an OrderDetail table containing information about the purchased product. Let's say I'd like to add a new column 'total' to skip calculation on the database every time I want to know the totals. It should be able to initialise the value by doing 'times ordered * price' for every existing row. Is that possible as well?
I did a seach here and found some posts but none that answered my specific problem.I am a programmer tasked with building an application for generating Quote Proposals. The database is for the most part fairly simple except when it comes to versioning and history. Basiclly every quote can be revised and modified several times prior to making a final decision (final approved quote), so I need to keep track of the changes that occur durning the revision process. I am not a DBA but I have had some database experience. From what I can tell I have two choices a.) Duplicate all the data everytime the quote is revised. While this method is does cause a lot of duplicate data it is very straitforward and easy to explain (or turnover to someone else) and reporting becomes very easy as well.Reporting.... this is my biggest area of concern as the users of the app should have the ability to print out the original quote proposals as well as the revised quote proposals. Duplicating all the data makes reporting very easy.b.) Create a history table and record the original data (along with who and when) before recording the new value in the main table. While this method does conserve disk space it make reporting a bit difficult as you would have to pull the specific value for a specific Quote Revision and display the orginal values on the report instead of the current ones.Table info: I looking a 10 to 12 tables to record and store the data. The largest table will have about 40 fields. Current estimates are producing about 5 to 8 quotes per week. Each quote is revised an average of 2 to 3 times.Are the pros and cons I listed the main ones to be concerned with and are there any other options?Thanks
I have a small webapplication, sql server 2000. The users can only update the data in the system. However my client needs a report that display changes. The changes are that needs to be monitored are only change of order status, change of delivery date and when a user splits an order.
What is the best practice to keep track of changes? A mirror table for each table with changes?
Hi All, I'm new on SSIS, but have worked for some time with DTS and a long time with other ETL tools like Informatica or OWB.
I would like to know in which way can i, easily, control my project/package versions. At the same time i need to implement a concurrency management system, which will control what developer is using which package, and when finished update the central repository (As it does Informatica or even OWB).
I have heard that i could implement versioning with source safe, but can i implement this in the way that i've referenced before. Can i use CVS?
For the past few months I've been developing an DW and ETL with SQL 2005 / SSIS. My packages are being deployed to a SQL Server. Although in the end game we will have a Dev/Staging/Production environments, I would still like to archive production packages when we push staging to production. Essentially I would like to archive the last X packages that were deployed to production where X is a reasonable number (3 - 5). I don't necessarily need to have them accessible to run. One of the purposes is to have another safeguard should we miss anything in user testing and need to roll back a deployment.
I am utilizing VSS and we will have backups running on the production server, but I would prefer to have a archive that is a little more accessible.
I just wondering if anyone has any thoughts on how to extract/archive production packages when the push is made. I could easily develop an app that queries the MSDB and exports the packages to the file system.
Edited:Maybe I should have posted this to the "managed" newsgroup. If any admins think that would be better, then let me know. I don't want to duplicate unnecessarily.
Hi,
We developed a custom Control Flow task for SSIS (2005, not yet had a lot of time to look at 2008 yet) and found that it does not handle versioning, or an uninstall and the resulting lack of an addressable component very gracefully.
Here is a typical scenario:
Baseline
Install component MyCustomTask 1.0 Create Project Save Project
Action 1
Uninstall MyCustomTask 1.0 and don't install the new version (a typical user scenario!) Open Project SSIS acts like the world has ended, especially if the user forgot to manually remove the item from the toolbox Fix:None, obviously, but it would be nice to be a bit more graceful and informative. Backdoor Toolbox fix: "Cleanse" the toolbox when it goes haywire by deleting the toolbox.tbd, in Documents and Settings<UserName>Local SettingsApplication DataMicrosoftVisualStudio8.0
Action 2
Uninstall MyCustomTask 1.0 cleanly, plus removing the toolbox item by hand. Install MyCustomTask 1.1, with identical interfaces etc, and add the toolbox item by hand. SSIS acts like the world has ended, and fails to ask you a sensible question like "do you want to upgrade the project to use the new version of the component" Fix:Identify major and minor version component changes and throw the user a rope. Backdoor Fix: Go into the DTSX manually - attack the DTSExecutable ExecutableType and DTS Name, for a Task in our case and replace it with the new version info. Even if the interface for the component has changed slightly, it seems to deal with that OK.
Given the fact that it seems to be very likely that there will need to be SSIS version specific builds of components (I am assuming that a task created in 2005 will not work with 2008), what is the best way to deal with the current lack of SSIS smarts.
Would this be the best approach:
Version the interfaces, but never the builds within a version i.e. My.CustomTask90 v1.0, My.CustomTask100 v1.0 etc.
This is a bit of a pain, rather than the simpler My.CustomTask v9.0 / v10.0 etc.
Or, are there some nice improvements in the pipe to alleviate this, plus perhaps even a way to programatically add components to the toolbox, rather than the low-rent method of getting the user to do it by hand.
in simple words it's about versioning at record level.ExampleTableEmployee - EmployeeId, EmployeeName,EmployeeAddress, DepartmentId,TableDesignationMap - EmployeeId, DesignationId, EffectiveDate,validityTableDepartment - DepartmentId, DepartmentTableDesignation - DesignationId, designationVia Modify-Employee-Details screen following are editableEmoyeeNameEmployeeAddressDepartmentDesignationthis screen should allow user to navigate through changes history.Example :Version -1EmoyeeName John SmithEmployeeAddress 60 NewYorkDepartment AccountsDesignation AccountantVersion -2EmoyeeName John SmithEmployeeAddress 60 NewYorkDepartment AccountsDesignation Chief Accountant - changedVersion -3EmoyeeName John SmithEmployeeAddress 60 NewYorkDepartment Sales - changedDesignation Marketing Manager - changedQuestion :What is the best proposed database design for maintaining historyrecords bound with version and retrieval techniqueBest RegardsSasanka
I have a question on locking pattern of read committed with snapshot isolation level that when two transaction update two different records then why do they block to each other even if they have previous committed value (old version of record).
I executed the below batch from a query window in SSMS
--Session 1: use adventureworks create table marbles (id int primary key, color char(5)) insert marbles values(1, 'Black') insert marbles values(2, 'White') alter database adventureworks set read_committed_snapshot on set transaction isolation level read committed begin tran update marbles set color = 'Black' where color = 'White'
--commit tran
Before committing the first transaction I executed below query from second query window in SSMS
--Session 2: use adventureworks set transaction isolation level read committed begin tran update marbles set color = 'White' where color = 'Black' commit tran
Here the first session blocks to second session. These same transactions execute simultaneuosly in snapshot isolation level. So my question is why this blocking is required in read committed with snapshot isolation level?
Hi all, Ther is a DSN (fileDSN, SystemDSN, UerDSN) on the client machine. User of the client machine has created a DSN which is connecting to the MS-Access database of his own machine. When user deploy that DSN to the Local Server, Server getting problem to finding the DSN (which is made on client). Please help me how to Deploy this DSN to server.
I do not have access to VS, but I hear it can do this... how can I do it without VS?
I am using CREATE ASSEMBLY to load dll form my server to SQLServer on my server (Same machine). How can I load the dll to another server? Do I have to write it on a directory on that server first? I tried to run the command from my pc connected to the other server, and I get a "cannot find file" error.
I would like to deploy several reports to production server, Do i need to install reporting services entire software in order to run the reports or is it possible to just have runtime files installed on it to run the reports.
please help, i have almost 100 reports to be deployed on this server which is located in other country.
Thanks for the helpful information.
(i am using SQL server 2005 / reporting services 2005.)
Problem: (asp.net 2.0)I x-copy from my development server a database named C:site_testsite.mdf to C:site. When I attach the database with Server Management Studio Express, it list the name as: "C:site_testsite.mdf" eventhough on the live server the database is in the C:site directory and my connection string in my web.config is "C:sitesite.mdf". But everything works until a week later when I need to make a field change to a table on the live site: I'm forced to re-attach the database, but this time it list the database as:C:sitesite.mdf". When I re-attach I get a NT AUTHORITYSERVICE does not have permission. So I give NT AUTHORITYSERVICE permission on the directory and files and ownership with Server Management Stuidio Express, but I still get the error.What am I'm doing wrong? Thanks for the help--Dietrich
I am using a login control ( using SQL express) for my application. Everthing works fine in my local machine, but after I deploy the application to the production server, I am getting an error says: An error has occurred while establishing a connection to the server. When connecting to SQL Server 2005, this failure may be caused by the fact that under the default settings SQL Server does not allow remote connections. (provider: SQL Network Interfaces, error: 26 - Error Locating Server/Instance Specified) I think it's becasue the SQL Express database which has the user login information has been set up correctly in the production server: Here is my connection string for SQL Express in my local box <add name="SQLExpress" connectionString="Data Source=.SQLEXPRESS;AttachDbFilename=|DataDirectory|ASPNETDB.MDF;Integrated Security=True;User Instance=True" providerName="System.Data.SqlClient"/> Any ideas what I should do to solve this issue
Can someone tell me the easiest way to deploy a SQL/MSDE server database via an installation package.
I have developed an application using SQL server and VB .Net. I am now writing the deployment/setup.exe that will create the database in SQL server. I will have a seperate client installation that will install the program files for client computers
I thought it would be as simple as connecting to SQL server and then executing the script that I had created using Enterprise manager Generate scripts wizard. This script runs OK in the query analyzer but not using conn.execute script.
I know there must be a simple solution that everyone else knows. What is it? Your help is appreciated !
i want to develop a web application that use SQL Server Database, i have installed SQL Server & VS.NET in my local computer and i'm ready to develop, but before getting started, i want to know how i will upload mydatabae in my host,
actually i want to know should i know the settings & properties of my host that support SQL Server, or without know them i can devekop my apploication and after developing localy with just a few tasks upload my ( Sql Server) database?????
i'm worry actually about this, plz say what you know about this. thanks
I'm on day 3 of troubleshooting.... still no luck!Very simple application. Extract data from the database and populate the GridView.Following error message:Server Error in '/WebSite3' Application.
An attempt to attach an auto-named database for file C:Webspaceigiriabcabc.comwwwWebSite3App_DataDatabase.mdf failed. A database with the same name exists, or specified file cannot be opened, or it is located on UNC share.-------------------------------------------------Look familar? so i Googled it and tried everything, from removing the User Instance = true in the web.config and even adding intial Catalog = Database and it didnt work.Instead i was greeted with a Newer Error msg:Server Error in '/WebSite3' Application.
Could not open new database 'Database'. CREATE DATABASE is aborted.Could not attach file 'C:Webspaceigiriabcabc.comwwwWebSite3App_DataDatabase.mdf' as database 'Database'.File activation failure. The physical file name "D:-- Work Documents --WorkezabcWebSite3App_DataDatabase_log.LDF" may be incorrect.The log cannot be rebuilt when the primary file is read-only.Next i have no idea why it's pointing back to my location directory hence that could be the issue but i'll leave it to you experts for your valueable input.It's uploaded onto a web server running ASP 2.0 and MS SQL Server 2005 i thinkFeel free to ask if you need additional informationThanks for reading!
Sorry for posting this here, I couldn't find an accurate forum name to post this in.
We are using Windows SBS Server 2003 with SQL Server installed. We have a static IP. I am wondering if we can host our database. I am specifically interested in having outside websites access our database. Is there anything special I need to do to make this happen? I am pretty clueless about this.
As we know Sql Server 2005 provides its ability to deploy clr assembly on it. However my problem is that my assembly has a configuration file. I used ConfigurationManager.GetSection to create some of the object. When I am deploying the project the trouble is that the app.config is missing. I used a class lib project type to create the assembly. Besides changing a storing method, can I still use this config file to reach my goal? Sqlserver.exe.config is not a good idea and it is not workable. The sqlserverexe host cannot find the assembly files.
