Staged-Rollout Loop

loop pattern · maturity: established · risk: high

Changes deploy in expanding stages (canary, cohort, full) with health verification and a rollback decision between each stage. The agent's job is running the checks and honoring the gates.

When to use

Agent-driven deployment where a bad release must be caught at 1% exposure, not 100%.

Steps

StepWhat happensVerification
Stage deployRelease to the current stage's scope only.Scope confirmed via the deploy system, not assumed.
Health verifyCheck the stage's health signals against pre-agreed thresholds for the soak period.Thresholds written before the rollout; signals read from real telemetry.
Advance or roll backHealthy advances to the next stage; unhealthy rolls back and escalates with evidence.The decision is recorded with the signal values that drove it.

Specification

Verification methodPre-agreed health thresholds on real telemetry per stage.
Recovery behaviorrollback-and-escalate
ObservabilityStage log with signals and decisions.
InputsRelease artifact, Stage plan, Health thresholds
OutputsSafe full rollout, or contained failure
Risk level / maturityhigh / established

Failure modes

  • Thresholds defined after seeing the data
  • Soak periods shorter than failure onset
  • Rollback path untested until needed

Assessment

Deployment engineering's proven loop, restated for agents because agents now run deploys. The rule that thresholds precede data is what makes the health check verification instead of rationalization.

Go deeper