我有一切必要的重新创建我的数据库从裸露的金属，减去数据本身。我相信有很多方法可以做到这一点，但我所有的脚本等都存储在subversion中，我们可以通过将所有这些从subversion中取出并运行安装程序来重建DB结构等。

我们有一个每周的sql转储到一个颠覆回购。这是完全自动化的，但这是一项非常繁重的任务。

你会想要限制修改的数量，因为它真的会在一段时间后消耗磁盘空间!

I agree with many of the posting concerning ruby's ActiveRecord migrations - they are an elegant way to manage the database in small incremental files that everyone can share. With that said, I've recently implemented a project using VisualStudio's Database Project, and it's kinda made me a believer. Short story - you create a database project, import all (if any) existing database objects into it (tables/views/triggers/keys/users/etc). That import results in a "Create" script per object. To manage the database you alter the create script and then on deploy VS compares the target database to the state of the database residing in your project and apply the proper alter statements.

这真的有点神奇，我必须承认，这是VS团队做的更好的事情之一。到目前为止，我真的很感动。

当然，您可以在自己选择的版本控制系统中管理整个数据库项目。

I've heard people say you absolutely have to keep your schemas in the database. I'm not sure I agree. This really depends on the system you're working with. If your system is relatively small and the data is not terribly important. And the the speed at which you need to bring another development environment online is crucial.. then yes.. you can benefit from it. However when your schema is useless without the data and the database is extremely large, it becomes virtually impossible to "source control" your database. Sure, you can still keep your DDL code in source control but that's essentially useless. You can't get the data needed without backup/restore.

在大型数据库开发工作中，我发现备份和恢复是首选的回滚选项。当然，你可以在源代码控制中保留过程、视图、函数等，但要保留表。SQL不是必需的。此外，如果您的部署过程是无懈可击的，那么您很可能永远不需要“回滚”您的生产环境。

作为一种规则，我们将所有的对象代码(存储过程、视图、触发器、函数等)都保存在源代码控制中，因为这些对象都是代码，而且几乎所有其他答案都同意，代码属于某种形式的版本控制系统。

As for CREATE, DROP, ALTER statements, etc. (DDL), we developed and use BuildMaster to manage the deployment of these scripts such that they can be run once and only once against a target database (whether they fail or not). The general idea is that developers will upload change scripts into the system and when it comes time for deployment, only the change scripts that haven't been run against the target environment's database will be run (this is managed very similarly to Autocracy's answer). The reason for this separation of script types lies in that once you manipulate a table's structure, add an index, etc., you effectively cannot undo that without writing a brand new script, or restoring the database - as opposed to the object code where you can simply drop a view or stored procedure then recreate it.

例如，当您将生产数据库恢复到集成环境中时，可以看到一些好处，系统会自动准确地知道哪些脚本没有运行，并将更改新恢复的数据库的表结构，使其与开发相关。

这对我来说也一直是一个很大的烦恼——似乎对开发数据库进行快速更改，保存它(忘记保存更改脚本)太容易了，然后你就卡在那里了。您可以撤消刚才所做的并重新做它以创建更改脚本，当然，如果您愿意的话，也可以从头开始编写它，尽管这将花费大量的时间来编写脚本。

A tool that I have used in the past that has helped with this some is SQL Delta. It will show you the differences between two databases (SQL server/Oracle I believe) and generate all the change scripts necessary to migrate A->B. Another nice thing it does is show all the differences between database content between the production (or test) DB and your development DB. Since more and more apps store configuration and state that is crucial to their execution in database tables, it can be a real pain to have change scripts that remove, add, and alter the proper rows. SQL Delta shows the rows in the database just like they would look in a Diff tool - changed, added, deleted.

一个很好的工具。以下是链接: http://www.sqldelta.com/