Power Platform: Planning a Controlled Rollback When a New Release Breaks Production
Diesen Beitrag auf Deutsch lesen
How to plan a controlled rollback in Power Platform: solution versioning, managed solutions, pipeline redeployment, Dataverse backups and separate flow recovery when a release disrupts production.
TL;DR
A failed Power Platform release needs a planned rollback, not a quick solution import. Classify the failure first: application-level issues may need a previous solution version, automation failures need flow and connection checks, and data corruption may require environment restore. Keep version-controlled solution artifacts, use managed solutions in production, enable redeployment of older pipeline versions, and take a manual Dataverse backup before major releases.
Original by Softchief Author, on Softchief Technologies. Read the original
This is our own summary, not a republication or full translation.
Governance takeaway
- Makers: Version every release, retain managed packages with deployment settings and release notes, and understand update, patch and upgrade behavior before trusting an import to reverse changes.
- Admins/CoE: Enable redeployment of older versions in Pipelines, restrict it to authorized staff, and take a manual environment backup before major releases, since system backups default to seven days.
- Security/Compliance: After a rollback or restore, recheck connection references, authentication and app sharing, because a restored environment turns solution flows off and leaves access unvalidated.
Frequently asked questions
Can Power Platform Pipelines roll back to a previous solution version?
Yes, with the Allow redeployment of older versions option, authorized users can redeploy a previous successful version from the pipeline run history.
Can a production Power Platform environment be restored over itself?
No, a production source cannot be restored directly over itself; it can be restored to a sandbox target, which is then placed in administration mode.
Why does a Power Platform rollback need a separate Dataverse backup?
A solution package holds customizations only, so a manual environment backup is needed to recover business records changed after deployment.
