Rules Configuration

Configure time-based schedules that automate ECS service runtime for linked AWS connections.

Overview

Rules tell Runless when to stop and restart services. They help you:

  • Reduce idle runtime in non-production environments
  • Standardize start/stop windows across teams
  • Keep manual override capability when needed

Rules are configured inside service or cluster configuration panels. There is no separate Rules screen.

Rule Model

Runless uses one rule model per service: a daily schedule with active days.

FieldDescription
Startup Time (UTC)Time of day to start the service
Shutdown Time (UTC)Time of day to stop the service
DaysDays of week when schedule is active
Desired Task CountTask count restored during startup window

Each service stores one active rule at a time.

Create a Service Rule

1

Open service panel

2

From dashboard services table, click the service name.

3

Open Configuration tab

4

Go to Configuration and enable cost saving.

5

Configure values

6
  1. Set Desired Task Count
  2. Set Startup Time (UTC)
  3. Set Shutdown Time (UTC)
  4. Select active Days
7

Review estimate and save

8

Check projected monthly savings and click Save.

Create a Cluster-Wide Rule

To apply one schedule across every service in a cluster at once:

  1. Select the cluster and open the Cluster-wide schedule panel above the service list
  2. Choose Configure schedule, define the active days and startup/shutdown times
  3. Choose Apply to all services

The same schedule is written to every schedulable service in the cluster. Afterwards you can still open an individual service and give it a different schedule - that per-service edit overwrites only that service. Re-applying the cluster-wide schedule overwrites all services again (including any you tuned individually), so apply the cluster baseline first and customize exceptions afterwards. If services currently have different schedules, the panel says so and an apply overwrites them all.

EC2 offers the equivalent at the region level: pick a region and use the Region-wide schedule panel to apply one schedule to every schedulable instance in that region.

Edit or Remove a Rule

Edit

  1. Open service or cluster configuration
  2. Adjust schedule values
  3. Save changes

Remove

  1. Open configuration panel
  2. Use Clear rule or Remove action
  3. Save changes

Removing a rule does not automatically restart stopped services. Use manual start when immediate restoration is required.

Execution Model

  1. Rule configuration is stored in the Runless tenant control plane
  2. Scheduler/worker flows trigger runtime actions at configured times
  3. Workers assume the linked account role and update ECS service desired count
  4. Status and history updates are written back to dashboard data

Best Practices

Keep schedules in UTC

Define schedules in UTC and review daylight saving impacts for your operating region.

Start from cluster defaults

Apply cluster-wide schedules first, then customize exceptions at service level.

Validate desired task count

Ensure startup task count aligns with autoscaling bounds and service dependencies.

Test before wider rollout

Validate new schedules on dev/staging workloads before applying broadly.

Troubleshooting

Save fails

Check:

  1. Startup and shutdown values are valid and different
  2. Desired task count is within allowed range
  3. Connection is verified and commercially active

Rules do not execute as expected

Check:

  1. Rule is still enabled
  2. Selected days and UTC times are correct
  3. Runtime history contains scheduler or action errors

Next Steps