使用 SQL Server 2008 将 TFS 2008 恢复移动到服务器
作为 TFS 2008 实例灾难恢复测试的一部分，我希望将数据库备份（当前在 SQL Server 2005 上）恢复到 SQL Server 2008。
我注意到 MSDN 上有一条警告，称这可能会使恢复变得不可能。
这是在野外进行的吗？ 有哪些风险？
As part of a disaster recovery test of our TFS 2008 instance I hoped to restore our backups of the database (currently on SQL Server 2005) to SQL Server 2008.
I noticed a warning on MSDN that this could make a restore impossible.
Has this been done in the wild? What are the risks?
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（2）
您的灾难恢复计划不应同时包含升级。 谨慎的做法是先进行升级，然后再规划灾难恢复计划。 等待恢复机会可能会带来麻烦。
也就是说，只要您使用最新的服务包，SQL Server 2008 现在就完全支持 TFS 2008。
Buck Hodges 是一位权威的 TFS 大师。 他是这样说的：
http: //blogs.msdn.com/buckh/archive/2009/04/09/tfs-2008-supports-sql-2008-sp1.aspx
Your disaster recovery plan should never include an upgrade at the same time. It's prudent to make the upgrade first and then plan your disaster recovery plan. Waiting for a recovery opportunity could spell trouble.
That being said, as long as you're using the latest service packs, TFS 2008 is fully supported on SQL Server 2008 now.
Buck Hodges is a definitive TFS guru. Here's what he says:
http://blogs.msdn.com/buckh/archive/2009/04/09/tfs-2008-supports-sql-2008-sp1.aspx
我对 TFS 2008 进行了一次升级，并尝试将备份恢复到 SQL Server 2008 实例。 我们还从 TFS 2008 的最终测试版进行升级，因此我不能说我们遇到的问题是由于测试版所致，还是由于 SQL Server 版本的更改所致。 这变成了一场小灾难，我们最终重新安装了全新的 TFS 并将我们的源代码复制到其中（幸运的是，它已在几台开发人员机器上进行了全面检查）。 我们在短时间内丢失了源代码控制历史记录，但最终能够重新启动并运行原始实例，并将其保留下来以供遗留源代码控制历史记录参考。
I have performed one upgrade of TFS 2008, and we tried to restore backups to a SQL Server 2008 instance. We were also upgrading from the final beta of TFS 2008, so I can't say whether the issues we encountered were due to it being from a beta, or if it was due to the change in SQL Server versions. It turned into a minor disaster, and we ended up reinstalling TFS fresh and copied our source code into it (which luckily was fully checked out on several developer machines). We lost our source control history for a short time, but we were eventually able to get the original instance up and running again, which we kept around for legacy source control history reference.