Automations That Stopped Working

Why Automations Break

Automations are built on top of everything else in your smart home stack. They depend on devices being online, sensors reporting accurately, hub software running correctly, cloud services being available, and timing systems being accurate. A failure at any level can cause an automation to silently stop working.

The challenge with broken automations is that they often fail silently. Unlike a device going offline, which usually triggers a notification, an automation that does not fire simply does nothing. You might not notice for days that your porch light stopped turning on at sunset or that your thermostat is no longer adjusting when you leave.

Diagnosing Trigger Failures

Every automation has a trigger: the event that starts it. When an automation stops working, the trigger is the first thing to investigate. Common trigger issues include:

• Sensor devices offline: If your automation is triggered by a motion sensor, contact sensor, or temperature sensor, check that the sensor is online and reporting data. A sensor with a dead battery does not trigger "no motion" or "door closed." It simply stops reporting, and the automation never fires.
• Time-based triggers with stale data: Automations triggered by sunrise or sunset rely on your hub knowing your location. If your location is set incorrectly or the time zone has drifted, these triggers fire at the wrong time or not at all. Verify your location settings and check that the calculated sunrise and sunset times match reality.
• State change vs. state: Many platforms distinguish between "when a state changes to X" and "while state is X." If you changed a trigger from one to the other, the behavior changes dramatically. A "state changes to" trigger fires once at the moment of change. A "while state is" trigger may need a periodic evaluation that your platform may not support.
• Device was replaced or re-added: If you removed and re-added a device, it may have a new internal ID. Automations that referenced the old device ID will silently fail because they are pointing to a device that no longer exists. Check your automation and update any device references.

Diagnosing Condition Failures

Conditions are filters that prevent an automation from running even when the trigger fires. If your automation's trigger is working but the actions never execute, a condition is probably blocking it.

Review each condition in your automation carefully. The most common condition issues are:

• Time windows that no longer apply: An automation set to run "only between 6 PM and 11 PM" will not run at 5:45 PM even if the trigger fires. Daylight saving time changes can shift your intended window unexpectedly.
• Mode or scene conditions: If your automation requires your home to be in "Home" mode and you forgot to update the mode when you arrived, the automation will not fire. Check your home's current mode or scene setting.
• Device state conditions: A condition like "only if the living room light is on" requires that specific device to be in the expected state. If someone manually turned it off, the condition blocks the automation.
• "And" vs. "Or" logic: If your automation has multiple conditions joined with "and," every single condition must be true. If even one is false, the automation is blocked. Double-check whether you intended "and" (all must be true) or "or" (any one is sufficient).

Diagnosing Action Failures

Sometimes the trigger fires and conditions pass, but the actions fail to execute properly. This is usually visible in automation logs if your platform provides them.

Common action failures include:

• Target device offline: The automation triggers correctly but the target device has gone offline. The command is sent but never received. Most platforms do not retry failed actions, so the command is simply lost.
• Rate limiting: Some platforms limit how frequently you can send commands to a device or how many automations can run per minute. If you have multiple automations firing simultaneously, some may be silently dropped.
• Conflicting automations: Two automations that control the same device can fight each other. For example, one automation turns a light on at sunset while another turns it off when no motion is detected. If both fire within seconds, the result depends on timing and is unpredictable.
• Service calls that changed: Platform updates sometimes rename or modify the available actions. An automation that used to call "light.turn_on" with certain parameters might fail after an update changes the parameter names or format.

Using Logs and Traces to Debug

Most smart home platforms provide some form of automation logging. Learning to read these logs is the single most valuable troubleshooting skill for automation issues.

In Home Assistant, the automation trace feature shows exactly what happened: which trigger fired, which conditions were evaluated (and whether they passed or failed), and which actions were attempted. Go to Settings > Automations, click on the automation, and view the trace. Each run is recorded with a timestamp and detailed execution path.

In SmartThings, check the automation history under the Automations tab. It shows recent executions and whether they succeeded or failed. For more detail, the SmartThings API and IDE provide device event logs.

In Alexa, check the Activity section in the Alexa app. Routine executions appear here with timestamps, though the detail level is limited compared to more advanced platforms.

Preventing Future Automation Failures

Build resilience into your automations from the start:

• Add fallback actions: If an automation controls something critical (like security lights), add a second automation with a different trigger that achieves the same result. Belt and suspenders.
• Use notifications for critical automations: Add a notification action to important automations so you know when they fire. If you stop getting the notification, you know the automation has stopped working.
• Test after every platform update: Manually trigger your most important automations after any hub firmware update, app update, or device firmware update. Five minutes of testing can prevent days of silently broken automations.
• Keep automations simple: Complex automations with many conditions and branches are harder to debug. When possible, break complex logic into multiple simpler automations that are easier to monitor and troubleshoot individually.