The problem
I have a light group with 3 Tuya and 2 Sengled bulbs. Tuya are wifi via cloud and Sengled are Zigbee. At random times when I run my automation to turn them on, the automation is triggered fine but occasionally one or more of the Tuya bulbs does not come on due to a network error causing the automation to abort. The message is that the automation stopped due to an error. Specifically "network error: (-9999999) sign invalid". If I follow up with another triggered automation or directly through the device, the offending bulb nearly always responds correctly without restarting. If it does not immediately work, it does after 30 seconds or so. Sometimes it happens when turning the commanding the bulb off and sometimes the problem interrupts more than on bulb from being controlled.
I've seen references to similar problems but not with this error message specifically. Some have suggested that Tuya local might be more reliable. Others suggest the contrary.
Is there a way to catch error in an automation and retry a step?
It seems like a hacky bandage but do I need to try the on or off command twice in every automation. Yuck!
In the screenshot the offending bulb is not listed.
This obviously points to the superiority of Zigbee devices in this situation but I'd like to avoid replacing all my Tuya bulbs that I had before I started this HomeAssistant journey. I am definitely avoiding any more going forward.
Thanks!

What version of Home Assistant Core has the issue?
core-2024.11.2
What was the last working version of Home Assistant Core?
No response
What type of installation are you running?
Home Assistant OS
Integration causing the issue
Tuya cloud
Link to integration documentation on our website
https://www.home-assistant.io/integrations/tuya/
Diagnostics information
No response
Example YAML snippet
alias: Laundry_button_toggle
description: ""
triggers:
- trigger: state
entity_id:
- binary_sensor.flic_80e4da7e0a1e
from: "on"
to: "off"
id: flic_pressed
- trigger: state
entity_id:
- input_button.laundry_lights_on
from: null
id: dashboard_button_pressed
conditions: []
actions:
- if:
- condition: state
entity_id: light.laundry_lights
state: "off"
then:
- action: light.turn_on
metadata: {}
data: {}
target:
entity_id: light.laundry_lights
- condition: trigger
id:
- flic_pressed
- action: timer.start
metadata: {}
data:
duration: "00:15:00"
target:
entity_id: timer.laundry_timer
else:
- action: timer.cancel
metadata: {}
data: {}
target:
entity_id: timer.laundry_timer
- action: light.turn_off
metadata: {}
data: {}
target:
entity_id: light.laundry_lights
- action: switch.turn_off
metadata: {}
data: {}
target:
entity_id: switch.bench
mode: single
Anything in the logs that might be useful for us?
Additional information
This is running directly on an older NUC. Works fine in general.
Devices are all using DHCP.
Host has dedicated IP
The problem
I have a light group with 3 Tuya and 2 Sengled bulbs. Tuya are wifi via cloud and Sengled are Zigbee. At random times when I run my automation to turn them on, the automation is triggered fine but occasionally one or more of the Tuya bulbs does not come on due to a network error causing the automation to abort. The message is that the automation stopped due to an error. Specifically "network error: (-9999999) sign invalid". If I follow up with another triggered automation or directly through the device, the offending bulb nearly always responds correctly without restarting. If it does not immediately work, it does after 30 seconds or so. Sometimes it happens when turning the commanding the bulb off and sometimes the problem interrupts more than on bulb from being controlled.
I've seen references to similar problems but not with this error message specifically. Some have suggested that Tuya local might be more reliable. Others suggest the contrary.
Is there a way to catch error in an automation and retry a step?
It seems like a hacky bandage but do I need to try the on or off command twice in every automation. Yuck!
In the screenshot the offending bulb is not listed.
This obviously points to the superiority of Zigbee devices in this situation but I'd like to avoid replacing all my Tuya bulbs that I had before I started this HomeAssistant journey. I am definitely avoiding any more going forward.
Thanks!
What version of Home Assistant Core has the issue?
core-2024.11.2
What was the last working version of Home Assistant Core?
No response
What type of installation are you running?
Home Assistant OS
Integration causing the issue
Tuya cloud
Link to integration documentation on our website
https://www.home-assistant.io/integrations/tuya/
Diagnostics information
No response
Example YAML snippet
Anything in the logs that might be useful for us?
Additional information
This is running directly on an older NUC. Works fine in general.
Devices are all using DHCP.
Host has dedicated IP