问题描述
我有两个数据工厂，我们称它们为test-dfact和prod-dfact.我已经在test-dfact中完成了更改(例如重命名管道)并导出了ARM模板.我在arm_template_parameters.json中相应地为生产编辑了参数 环境，然后我启动了导入ARM模板"来自prod-dfact.部署成功，但是prod-dfact现在拥有它以前做过的所有资源，以及ARM模板中的所有新功能.特别是，我在test-dfact中重命名了管道 但它只是作为prod-dfact中的新管道添加的.
I have two data factories, let's call them test-dfact and prod-dfact. I have finalised my changes (e.g. renamed pipeline) in test-dfact and exported the ARM Template. I edited the parameters accordingly in arm_template_parameters.json for the production environment, and I have then initiated "Import ARM Template" from prod-dfact. The deployment succeeds, but prod-dfact now has all the resources it did before, plus anything new in the ARM Template. In particular, I renamed a pipeline in test-dfact but it just got added as a new pipeline in prod-dfact.
将这种类型的部署部署到生产环境的建议做法是什么?在部署具有dev/test更改的ARM模板之前，我是否应该删除整个生产数据工厂?
What are the recommended practices for this type of deployment to a production environment? Am I supposed to delete the entire production data factory prior to deploying the ARM Template with changes from dev/test? That doesn't seem wise to do in production!
推荐答案
我相信您会看到ARM增量模式的行为，这是ARM的默认模式.看一下这篇文章，其中解释了不同的模式以及您可能会发现有用的其他最佳实践:
I believe you are seeing the behavior of ARM incremental mode, which is the default mode for ARM. Take a look at this article which explains the different modes, as well as other best practices you might find useful:
https://blogs.msdn.microsoft.com/mvpawardprogram/2018/05/01/azure-resource-manager/
https://blogs.msdn.microsoft.com/mvpawardprogram/2018/05/01/azure-resource-manager/
这篇关于ADFv2-通过ARM模板部署到现有生产数据工厂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！