Autoscaling policy misconfigured: pre-release and first 30 minutes checklists

A scaling change fails to add capacity or scales to zero under load.

The pre-release checklist maps each item to a test class from the incident page. The first-30-minutes checklist covers detection, containment, the rollback decision and customer communication.

7 pre-release items · 7 first-30-minutes items · verified 2026-10-22

Pre-release checklist

PR-01Run a pre-production load test at peak-hours request volume against every change to the API auto-scaling rule and confirm the instance count rises above the off-hours baseline. load
PR-02During that load test, confirm API latency stays inside its budget while the new scaling threshold is in effect. load
PR-03Block the release of any scaling rule change that has no passing pre-production load test result. load
PR-04Compare every capacity threshold value in the autoscaling change with the live scaling state and confirm none is outdated. config
PR-05Confirm the minimum instance baseline in the new scaling rule matches peak-hours volume for the time of the deployment. config
PR-06Apply a scaling rule change to one cluster or a blue/green copy first and compare its serving instance count and latency with the unchanged cluster. canary
PR-07Configure the rollout to revert automatically when serving instances fall to the minimum baseline during business hours. canary

First 30 minutes

FM-01When latency or job-start alerts fire, compare the serving instance count with the minimum baseline.
FM-02Check whether the last deployment changed an auto-scaling rule or a capacity threshold value.
FM-03Roll back the scaling change before debugging it, because rolling back or reverting the change restored service in three of the five public reports.
FM-04If the rollback is slow, raise the instance count by hand to the peak-hours level.
FM-05Watch the queued backlog of workflow runs drain and declare recovery only after it clears.
FM-06Post a status update naming the affected cluster or service and the share of failed or delayed jobs.
FM-07Freeze further scaling configuration changes until a load test of the corrected rule passes.

Download the checklist

Pre-release checklist (Markdown)1.6 KB · v1.0.0
First 30 minutes checklist (Markdown)1.1 KB · v1.0.0

Only in this checklist

Every test class in this checklist also appears in at least one sibling checklist in the group; the item texts differ: Run a pre-production load test at peak-hours request volume against every change to the API auto-scaling rule and confirm the instance count rises above the off-hours baseline.

Questions

What should be checked before a release to prevent Autoscaling policy misconfigured?

7 pre-release items, PR-01 to PR-07. PR-01: Run a pre-production load test at peak-hours request volume against every change to the API auto-scaling rule and confirm the instance count rises above the off-hours baseline.

Is there a rollback plan for Autoscaling policy misconfigured?

The first-30-minutes checklist has 7 items. FM-01: When latency or job-start alerts fire, compare the serving instance count with the minimum baseline.

What should a team do in the first 30 minutes after Autoscaling policy misconfigured is detected?

The first-30-minutes checklist has 7 items. FM-01: When latency or job-start alerts fire, compare the serving instance count with the minimum baseline.

Which configuration checks apply to Autoscaling policy misconfigured?

PR-04: Compare every capacity threshold value in the autoscaling change with the live scaling state and confirm none is outdated. PR-05: Confirm the minimum instance baseline in the new scaling rule matches peak-hours volume for the time of the deployment.

Is a Autoscaling policy misconfigured checklist still needed if deployment is automated?

Yes. Automation runs the steps; the checklist names what the tests must cover: load and performance tests, configuration and environment tests and canary and progressive delivery checks.

Which pipeline stages does the Autoscaling policy misconfigured checklist cover?

The pre-release items map to the pre-deploy and post-deploy stages.