How to test fabric CI/CD Azure DEV ops ?

[kkoc3]

Hi,

I've been assigned a task to validate whether our CI/CD pipeline for deploying Microsoft Fabric items via Azure DevOps will function correctly in a disaster recovery scenario. Essentially, this is a form of DR testing, and I'm trying to determine the best way to approach it.

Initially, I considered creating new items (e.g., Lakehouses, notebooks) in the Dev workspace that's linked to Azure Repos, then performing changes and rollbacks to simulate recovery. However, my manager advised against this approach — though I'm not entirely sure why.

My second idea was to clone existing items into a new workspace and run similar tests there. Unfortunately, that option was also declined.

I'm now looking for alternative ways to test the CI/CD process effectively. Specifically:

• What aspects of the pipeline should I be validating?
• Are there best practices or recommended approaches for DR testing in Fabric + Azure DevOps environments?
• How can I simulate failure and recovery without impacting production or violating workspace constraints?

Any insights or suggestions would be greatly appreciated.

[KevinChant]

It should work depending on how you work with variables and parameters in your pipelines in Azure DevOps

[v-nmadadi-msft]

Hi @kkoc3 

As we haven’t heard back from you, we wanted to kindly follow up to check if the suggestions  provided by the community members for the issue worked. Please feel free to contact us if you have any further questions.

Thanks and regards

[v-nmadadi-msft]

Hi @kkoc3 ,

May I check if this issue has been resolved? If not, Please feel free to contact us if you have any further questions.

Thank you

[kkoc3]

Hi Its not resolved yet as non of the comments have answers my question . I would like to keep this open

[tayloramy]

Hi @kkoc3, 

So far we've given examples of CI/CD frameworks, and BC/DR frameworks that have been validated bu the Fabric CAT team. 

What specific answers are you still looking for? 

If you found this helpful, consider giving some Kudos. If I answered your question or solved your problem, mark this post as the solution.

[kkoc3]

Hi,

I'm exploring how to effectively test our CI/CD processes, especially in scenarios that go beyond standard BCDR tooling.

Our CI/CD setup is quite similar to what you've described in your organization. Now, consider a scenario where someone accidentally deletes key assets in your Fabric environment — such as pipelines, dataflows, notebooks, etc. You're tasked with restoring everything.

I'm interested in understanding how you would approach this recovery. What processes or strategies would you follow to bring everything back?

I'm not looking for tools or features like geo-replication or region failover  this isn't about BCDR tooling.

Instead, I'm focused on processes and approaches that help validate the resilience and recoverability of your CI/CD setup.

This is just one scenario — I’m looking to explore a few others as well. The goal is to build a framework for testing and validating CI/CD robustness in real-world failure situations.

Does that make sense?

[v-nmadadi-msft · Accepted solution]

Hi @kkoc3 ,
Thanks for reaching out to the Microsoft fabric community forum.

For disaster recovery testing in Fabric with Azure DevOps, it’s best to avoid running tests in your Dev or Prod environments. Instead, create a dedicated sandbox workspace for example, DR-Validation-Workspace  and target your pipeline deployments there. This isolated environment allows you to validate the full recovery process end-to-end without risking disruption to active workloads.

I hope this information helps. Please do let us know if you have any further queries.
Thank you

[v-nmadadi-msft]

Hi @kkoc3 ,

I wanted to check if you had the opportunity to review the information provided. Please feel free to contact us if you have any further questions.

Thank you.

[kkoc3]

Hi 

yes but it doesnt answer my question . 

[tayloramy]

Hi @kkoc3, 

There are also CI/CD accelerators: 
https://github.com/microsoft/fabric-toolbox/tree/main/accelerators/CICD

The way my organization does CI/CD is as follows: 

We have a dev, test, and prod workspace. Our dev workspace is git enabled and we use feature branching out to new workspaces to do feature development, and then merge back to main once the feature is complete. Main is the branch in the dev workspace. 

Our test and prod workspaces are not git enabled, instead we use deployment pipelines to move from dev to test and then to prod. 

For DR and failure recovery, there is no built in fucntionality. This is why I linked the BCDR accelerator, that is the best starting point right now. 

Note these accelerators are not an official product and don't have support, but rather they are a starting point for you to build a full solution. 
If you found this helpful, consider giving some Kudos. If I answered your question or solved your problem, mark this post as the solution. 

[tayloramy]

Hi @kkoc3, 

The Fabric CAT team has actually released a BCDR accelerator in the Fabric Toolbox. I recommend checking it out! 
https://github.com/microsoft/fabric-toolbox/tree/main/accelerators/BCDR

Note this is not an official product and doesn't have support, but rather it is a starting point for you to build a full solution. 
If you found this helpful, consider giving some Kudos. If I answered your question or solved your problem, mark this post as the solution.