Write a deployment rollback procedure

“Redeploy the old version” is incomplete if the version, traffic action, migration compatibility and verification are unknown.

Inspect a runbook record

The integration lab creates an authored rollback procedure.

from integration_cases import rollback_case

result = rollback_case()
assert result["has_trigger"] is True
assert result["target_pinned"] is True
assert result["verification_steps"] == ["verify readiness", "check error rate"]
assert result["executed"] is False

Define metric/incident trigger, decision authority, current and target artifact digests, traffic or feature-flag action, database compatibility, cache/jobs treatment and communications. Preserve evidence before teardown.

Rollback must be tested in staging or a safe environment. If schema changes are not backward compatible, use expand/migrate/contract or a forward-recovery plan. After rollback, verify health, error rate and the original user scenario.

The FDE for Freshers course treats rollback as part of deployment evidence.

Exercise

Deploy two local container versions, inject a health regression and exercise the runbook. Record recovery time and every manual assumption.

Continue learning

This article is part of the FDE integration and deployment foundations sequence. Use the neighbouring tasks when you need the prerequisite or the next application.

• Review the prerequisite or neighbouring task in Connect an agent to a read-only internal tool.
• Continue with Hand over a student integration project with operational notes.

Reference: Kubernetes deployment rollback.