Terraform Academy · Production Runbooks

Capacity and Performance Degradation

Handle saturation, throttling and latency before full outage.

Objective: Recover performance while identifying the limiting resource.

Practical guidance

Signals

01

Latency percentiles

Apply this point to a real Terraform Academy design, delivery or operational scenario and record the evidence used to validate the outcome.

02

Queue depth

Apply this point to a real Terraform Academy design, delivery or operational scenario and record the evidence used to validate the outcome.

03

CPU and memory pressure

Apply this point to a real Terraform Academy design, delivery or operational scenario and record the evidence used to validate the outcome.

04

Dependency throttling

Apply this point to a real Terraform Academy design, delivery or operational scenario and record the evidence used to validate the outcome.

Practical guidance

Response

01

Reduce load safely

Apply this point to a real Terraform Academy design, delivery or operational scenario and record the evidence used to validate the outcome.

02

Scale validated bottleneck

Apply this point to a real Terraform Academy design, delivery or operational scenario and record the evidence used to validate the outcome.

03

Protect critical traffic

Apply this point to a real Terraform Academy design, delivery or operational scenario and record the evidence used to validate the outcome.

04

Capture before-and-after metrics

Apply this point to a real Terraform Academy design, delivery or operational scenario and record the evidence used to validate the outcome.

Practical guidance

Prevention

01

Capacity forecast

Apply this point to a real Terraform Academy design, delivery or operational scenario and record the evidence used to validate the outcome.

02

Load testing

Apply this point to a real Terraform Academy design, delivery or operational scenario and record the evidence used to validate the outcome.

03

Autoscale boundaries

Apply this point to a real Terraform Academy design, delivery or operational scenario and record the evidence used to validate the outcome.

04

SLO-based alerts

Apply this point to a real Terraform Academy design, delivery or operational scenario and record the evidence used to validate the outcome.