We run two Rachio controllers on a Fixed schedule with all Intelligence features turned on. We had a power outage lasting about 8 hours, during which it was pouring rain. The outage began before the time at which Rachio normally makes the decision whether to run the schedule, and spanned the time during which the schedule usually runs (if it is going to run).
Upon restoration of power (well after the normal start time), both controllers immediately began watering. What is odd about this is that they appear to have done so without checking weather conditions. Usually, the Schedule Updates log contains an entry indicating that schedules were skipped or not skipped according to observed/forecast weather conditions. No Schedule Updates log entry was recorded in this instance.
The power outage also took out our Internet service, but it came back online shortly after power was restored. Perhaps when power came on, the controllers discovered that they could not immediately communicate with the Rachio mother ship or with weather stations. But it still seems odd that the controllers decided immediately (within one minute of power-on) to run the schedules, without logging any entry in Schedule Updates about why it made this decision.
We have two controllers that behaved the same way, so this isn’t just some random transient glitch.
I just want to clarify- they began running at a time not even close to your scheduled start time? Did this happen with both controllers at the same time, or did this happen to you on two different occasions?
This happened last night for both controllers. Here is the timeline:
Power failed at around 6:30 pm.
The scheduled start time was 8:30 pm for one controller and 10:30 pm for the other.
Power was restored at 2:13 am. Both controllers logged a “Device status power cycle”. Both controllers then immediately cycled through all of the zones for the previously scheduled program.
At least, this is what the Watering History tells me! I was not out and about at 2:13 am and I wouldn’t have been able to tell if the sprinklers were actually running in the pouring rain. As I look at the Watering History now, I see something curious:
Do you see anything odd about that? Each of the zones completed in zero time. Also, there were no Started and Completed events recorded in the Watering History, as there are normally.
I’m now thinking that this is just an anomaly in the Watering History, rather than inappropriate watering as I thought. Sorry for the false alarm.
@eataft, sounds like you experienced the rare occasion in which your power and WiFi were out. Anytime the controller goes offline, it will run the last saved schedule without checking for weather (since it cannot receive updates from the Rachio cloud). For more information, please see this support article.
This is normal behavior as the controller will not delay your schedule start time until the controller comes back online, with the exception of Flex Daily schedules which are dynamic and will default to every 3 days after the controller has been offline for 24 hours.
Since the controllers would have ran the Fixed Schedules at the scheduled start time regardless of the WiFi connection (provided power was available to activate the solenoids) you’re seeing the event logs from the controller’s memory uploading after the controller reconnected to WiFi.
Since this is the same time stamp as when the controller power cycled and reconnected to WiFi, all events that took place while the controller was offline will share the same time stamp.
Aha! If it is true that events are time stamped according to when they were reported to the cloud, rather than when they actually occurred, then it explains everything that I observed.
I didn’t mention that my home has emergency power, but for only some circuits; it comes on when an outage lasts more than a few minutes. When I saw the reported behavior from Rachio, which seemed to indicate power-on at the time the outage ended, I figured that the controller must be plugged into the wrong circuit. But now I have double-checked and the circuit does receive emergency power.
So, what really happened is that the controller lost power only briefly and came on again in time to run its regular schedule. However, it had no Internet connection, so it ran the schedule without regard to weather. (The loss of Internet connectivity is Comcast’s problem, not mine; my modem and router are on UPS and did not lose power.)
I will enter a feature request to time stamp events according to when they actually occurred. The current behavior is very confusing when trying to figure out what happened.
