← Back to help
Troubleshooting
Start with the quick checklist below. Most problems are caused by a small setting mistake.
Start with these checks
- Device online? Can you open the device web UI from your home network (by IP)?
- MQTT enabled? Is MQTT enabled and does it show ?Connected??
- Correct credentials? Host/port/user/password copied from Optimaatti MQTT credentials page.
- TLS/SSL matches port? If TLS is enabled use TLS port; otherwise use non-TLS port.
- Unique clientId? Same clientId on multiple devices can break connections.
Decision tree: what problem do you have?
- Device does not appear at all: usually MQTT credentials, TLS/port or network.
- Device appears but is offline: connection drops, wrong port/credentials, or internet issues.
- Device online but does not react: commands not reaching device or rules overriding.
- Predictive control / Learning behaves oddly: sensor placement, data gaps, tight limits or wrong power.
- Offline scheduling not working: device does not support it, no schedule created, or sync failed.
Device does not appear in the system
Follow these steps to resolve the issue
1
Check MQTT settings
- Open the device web UI
- Go to Settings → Advanced → MQTT
- Ensure MQTT is enabled
- Copy host/port/user/password from Optimaatti MQTT credentials page
- Ensure TLS/SSL setting matches the chosen port
- Keep clientId unique
2
Check network connectivity
- Ensure the device has internet access
- Try pinging the device (e.g. ping 192.168.1.100)
- Ensure firewall/router does not block outbound MQTT
3
Restart the device
- Power off for 10 seconds
- Power on
- Wait ~30 seconds
- Check if the device appears in the dashboard
Predictive control does not behave as expected
Follow these steps to resolve the issue
1
Check target temperature and limits
- Open device settings
- Ensure min/max are realistic (e.g. 21–23°C)
- Too narrow range can cause frequent toggling
2
Check device type and sensor
- Ensure correct device type is selected
- Ensure temperature readings are realistic
- Verify sensor placement (no sun/drafts/near heater)
3
Allow the system to learn
- During the first days predictive control may need time to adapt
- Monitor for a few days before conclusions
4
Check price data
- Verify spot prices update normally
- If prices are missing, predictive control may fall back to a backup plan
Device shows as offline
Follow these steps to resolve the issue
1
Check device status
- Open device web UI
- If you cannot access it, the device is truly offline
- Check power and network cable
2
Check MQTT connection
- Open MQTT settings
- Check connection status
- If disconnected, verify server address and credentials
3
Regenerate MQTT password
- If it used to work, regenerate password in Optimaatti
- Update the new password in the device MQTT settings
Device does not react to commands
Follow these steps to resolve the issue
1
Check status
- Ensure device is online
- Refresh the page
- Ensure you have permission to control the device
2
Try manual control
- Open device web UI
- Toggle on/off there
- If it works, issue is MQTT; if not, issue is the device
3
Check rules
- A rule may override manual control
- Review active rules
- Temporarily disable rules to test
Offline scheduling not working
Follow these steps to resolve the issue
1
Ensure offline scheduling is enabled
- Open device settings in the dashboard
- Enable offline scheduling
- Save
2
Check device support
- Not all models support offline scheduling
- Supported: Shelly Plus and Pro
- Gen1 does not support it
- Requires schedule generation and successful sync via MQTT
3
Update firmware
- Open device web UI
- Go to Settings → Firmware
- Update to latest
- Restart after update
Price data does not update
Follow these steps to resolve the issue
1
Check ENTSO-E API
- Prices come from ENTSO-E API
- API may be temporarily down
- Wait 15 minutes and try again
2
Check timezone
- Open Settings → General
- Ensure timezone is correct (Europe/Helsinki)
- Save