具有完全枚举提供程序同步的初始数据SnapShot [英] Initial Data SnapShot with Full Enumeration Provider Sync
问题描述
我有两个自定义完整枚举提供程序似乎运行速度相当慢，我想通过在远程提供程序的MS同步之外创建一个快照来加速初始同步，以供本地使用。不幸的是，在执行此操作后，当我执行初始同步时，Microsoft同步尝试插入已存在的记录。 我假设这是因为单独的SQLMetaData存储不知道记录已经存在。 是这种情况还是还有其他问题？ 我认为在双方完全枚举期间，如果元数据匹配，它仍然不会进行插入。
dosborn，
您是如何创建快照的？我假设你只复制了数据。在内部，每个项目都有一个全局ID，它在整个同步生态系统中是唯一的，在您的情况下，Sync Engine认为它处理的2个DISTINCT项目恰好具有相同的数据。为了实现您想要的结果，您还需要传播元数据，以便引擎知道这些项目实际上是相同的。然而，根据经验，"触摸"是指"触摸"。元数据不是人们可以做的最好的事情，因为你最终可能会丢失数据。
除非这不是最高优先级，否则我建议等待初始同步完成它自己的。自填充所有数据以来，初始同步总是花费更多时间。以下同步（增量）更快。
希望这会有所帮助。
阿德里安
I have two Custom Full Enumeration Providers that seem to be running fairly slow and I wanted to at least speed up the initial sync by created a snapshot outside of MS sync of the remote provider for the local to use. Unfortunately after doing this, when I do the initial sync Microsoft sync tries to insert records that are already there. I assume this is because the separate SQLMetaData store is unaware that the records are already there. Is this the case or is there something else that is wrong? I would think that during the full enumeration of both sides, if the meta data matched it would still not do the insert.
dosborn,
How did you create the snapshot? I assume you copied the data only. Internally each item has a Global Id, which is unique across the whole sync ecosystem, and in your case the Sync Engine thinks it deals with 2 DISTINCT items which happen to have the same data. In order to achieve the result you wanted to, you'll need to also propagate the metadata, so that the engine is aware that these items are in fact the same. However, as a rule of thumb, "touching" the metadata is NOT the best thing someone can do, as you can end up losing data.
Unless this is not a top priority, I'd recommend waiting for the initial sync to complete on its own. Initial sync always takes more time since it populates all the data. The following syncs (incremental) are much faster.
Hope this helps.
Adrian
这篇关于具有完全枚举提供程序同步的初始数据SnapShot的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！
