Quiz: Edge & IoT Infrastructure

← Back to quiz index

3 questions

L0 (1 questions)

1. What is the A/B partition scheme for OTA updates, and why is it essential for edge devices?

Show answer

The A/B partition scheme keeps two OS partitions: one active and one for updates. Updates are written to the inactive partition. On reboot, the device tries the new partition. If it fails to boot (3 consecutive failures), it automatically reverts to the previous partition. This is essential because edge devices cannot be physically accessed for recovery if an update bricks them.

L1 (1 questions)

1. Why is standard Prometheus scraping unsuitable for edge devices on metered cellular connections, and what are two alternative strategies?

Show answer

Standard Prometheus pull-based scraping requires constant network availability and generates significant traffic that is expensive on metered cellular. Alternatives:
1. Edge-local Prometheus with remote_write that batches and compresses metrics before pushing to a central server.
2. MQTT-based metrics where the edge device publishes tiny payloads to a broker, which feeds into Telegraf and then Prometheus. Both reduce bandwidth and handle intermittent connectivity.

L2 (1 questions)

1. You manage 500 edge devices on cellular. An update bricked 3 devices in your canary group. What update safety rules should have prevented wider damage, and what is the correct phased rollout pattern?

Show answer

Safety rules:
1. Never update all devices at once — use canary deployment.
2. Every update must have automatic rollback (boot counter: if new version fails 3 times, revert; watchdog timer: if device does not phone home within N minutes, revert).
3. Never update the bootloader unless absolutely necessary. Correct phased pattern: 1% canary with 24-hour soak, then 5%, then 25%, then 100%, with observation and health checks at each phase before proceeding.