Outdated Data due to Cloud outage?

Help & Support
1

We finally got rain and I installed my two new units this past weekend. One seems to be working just fine. The other is not. I’ve attached a screen shot of an “Outdated Data” message that I get every day. It suggests there might be a cloud outage. But all my other units are working, so how can it be a cloud problem. Also, please note that this unit does not even show that it is in training mode.

What should I do? Delete this unit and start over??
Serial # 334RPU Manufactured date April 14, 2025

P25 N Highbush_Geodrops_20250827_msg
P25 N Highbush_Geodrops_20250827_msg720×1600 90 KB

2

Hi @mg.marsha.lehman ,

Can you try to refresh the data by going back to the home dashboard of all your devices and swiping down on the screen? I will flag engineering if this problem persists after this.

Stanley
Marketing PM
Team GeoDrops

3

Stanley:
I have done the refresh probably every other day. At your request, I just did it again.

I still have the message that the device has been “learning since Aug 24”.

But notice in the attached screen shot that the “last sent by device is Aug 24 at 3:30pm”.

P25 N Highbush_Geodrops_20250827_not learning
P25 N Highbush_Geodrops_20250827_not learning720×1600 85.1 KB

4

Hi Marsha,

I just checked your device. It looks like the your device last communicated with GeoDrops Cloud on 2025-08-26 at 23:54 (UTC), and this was before the device is installed in soil. My best guess is that the location that you installed this device actually has a very poor Wifi reception so that the device wasn’t able to connect to the Internet?

Thanks,
Lawrence

5

No. It was installed in the soil on Aug 24. After 48 hours, it had not left the “sensing soil” state, so I moved it (Aug 26).

6

Hi @mg.marsha.lehman ,

Thanks for the additional details. @homedigy-alpha will get back to you.

Stanley

7

Hi @mg.marsha.lehman

Something is definitely not right with the device. It basically stopped contacting the server since Aug 28. Can you check if the device is still online? If not, since power cycling the device would probably take as much effort as setting up the device again, perhaps it would be easier to just delete the device and reinstall it.

Please let us know if the device still behave incorrectly

Thank you for your patience

Al

Engineering

8

If I had responded immediately to your message, I would have said “yes, the device is still on-line”.

HOWEVER, when I checked the device today (after a major thunderstorm), I found that the device has gone OFFLINE again.

Tomorrow, I will do as you suggest, and delete the device and start all over with a fresh install.

Marsha

9

Hi @mg.marsha.lehman

I did notice that the device seems to have difficulty contacting the server. Is it possible for you to put the device at a location closer to your wifi AP? I would like to verify if it is only an internet connection issue that was causing the device to behave like this

Thank you

Al

Engineering

10

Al:
I recently installed two of the 2025 production units. This finicky one is closer to our WiFi than the other one. And the other one has worked well since it was installed on Aug 24.

More importantly, I really want this unit in this particular raised bed (which is wood, not metal).

I will first try your prior recommendation: delete the device and start over. If that doesn’t work, then I will test it in a different location.

Marsha

11

Al:
I deleted and reinstalled the device (serial #334RPU) today – or at least I spent a lot of time trying!

I had the batteries in and out at least 6 or 7 times. It would get to the point of connecting / configuring the device and I would get a message of “something went wrong, please try again”. So, I did. This happened 5 or 6 times. I then took the batteries out and just let the device sit in my garage while I had lunch.

The next time seemed to be the charm. I made it through the setup, and the device went into soil sensing mode. While the message says this will take 90 minutes, it has now been over six hours, and the device is still in soil sensing mode.

As we are approaching dusk, I’m going to let it be overnight. If it is still in soil sensing mode in the morning, I will try one more time. After that, I will declare this device defective.

For the record, this finicky device was manufactured on April 14, 2025. My second device, which is working fine, was manufactured on MAY 14, 2025. Is it possible I got a unit that just barely passed QC?

Marsha

12

Hi @mg.marsha.lehman

We will look into this over the next couple of days to verify. Thank you for your help and patience

Al

Engineering

13

Al:
Add this to the list of “interesting” things that have happened with this device (#334RPU).

As I mentioned earlier, I deleted and reinstalled this device – and finally got it connected. It went into soil sensing mode, which normally takes 90 minutes. As last evening’s post documented, six hours later it was still in that mode.

Well, this morning, now at the 18 hour mark, it was still in soil sensing mode. So, I removed the device from the soil, took it into the garage, cleaned the outside, opened the battery compartment, and removed the batteries. That was over an hour ago. But the device has not figured out that it is out of the soil, has no batteries, shows a good connection, and is still in soil sensing mode.

Here’s a screenshot of what I see while this device is in my garage (not in soil), and has no batteries in it.

Geodrops_soil sensing when not in soil_20250906
Geodrops_soil sensing when not in soil_20250906720×1600 148 KB

Marsha

14

Update: Sunday afternoon, 9/7/2025, 4:16pm

The system finally (after more than 24 hours) figured out that the device was offline. I took this as the opportunity to start again, and installed the sensor in a place that is as close as one can get to our internet hub. It is literally in an ornamental bed next to a window of the room where our Wi-Fi hub is located. This is for testing purposes only; I have no desire to put a sensor in this ornamental bed for longer than this test.

If the device doesn’t work in this location, I can’t imagine what to do next.

15

Hi @mg.marsha.lehman

After some digging on our end, we think the issue is a combination of poor wifi at your old setup location, intermittent update to our backend and confusing error messages as a result of said intermittent update. Our system updates different information at different intervals.

  • For moisture data it is usually every three hours, unless the Droplet registered a significant change in the soil, then it would do an immediate update.
  • For device status, it is around every 6 hours, unless the Droplet has urgent issues to report, such as HW problems.

Let’s go over each issue you encountered and what we think was the likely cause.

When this happened:

configuring the device and I would get a message of “something went wrong, please try again”. So, I did. This happened 5 or 6 times. I then took the batteries out and just let the device sit in my garage while I had lunch.

Is it correct to assume you were setting up the device close to where you were planning to put the device the first time? The initial reported wifi strength on our was quite weak so it likely took multiple retries for the device to be able to contact our backend.

image
image450×1000 107 KB

The “outdate data due to cloud outage” message was due to the device having difficulty uploading moisture data to the server at the old location. As mentioned before, the Droplet updates the device status and moisture data separately. The “last sent by device” was confusing because it was referring to the last time the server received moisture data, not the device status update.

Geodrops_soil sensing when not in soil_20250906

Geodrops_soil sensing when not in soil_20250906720×1600 148 KB

Due to the intermittent nature of the connection, even the status update of the device was not always up to date, which was why to the backend, you were still “stuck” in soil sensing at that point.

Long story short, the current test location where you have setup the device is great and the device seems to be working correctly. We will tweak the App messaging, notifications and update timestamp to make these situations less confusing for the user.

For the original location, is it possible to find a location closer to our wifi AP? For now, using the phone’s wifi signal bar may help you find a location with better wifi. We are looking into having more indicators in the app to do something similar in the future.

Thanks you for your patience Marsha and keep us posted on any new issues you may find.

Al

Engineering

16

Yes, the “something went wrong” message was generated in the “old” location.

While the device seems to be working ok in the “new” location, this location is about 10 feet from our wi-fi hub. More importantly, this is not a location where I want to have a GeoDrops sensor. I really want it back in the “old” location.

So how much of this was due to distance from the wi-fi vs. the confusing messages and backend updates on your side.

What I find interesting is that I have a second unit (also 2025 production) that is further away from the wi-fi than this unit had been – and that unit appears to be working just fine.

I also find it inconvenient to have to move the device and then go through two weeks of training – and then have to move the device to where I want it and go through two more weeks of training. That chews up a month of my gardening season, and is a period in which I have little (if any) usable data.

Marsha

17

Hi @mg.marsha.lehman

Can you let us know which device you are referring to. We would like to verify if the device in question’s wifi is functioning normally

Thanks

Al

Engineering

18

The unit in question, and that I have moved to within 10 feet of our wi-fi hub is serial number 334RPU. (This was documented in an earlier post.) This is a 2025 production unit, with a manufacture date of 4/14/2025.

I have a second 2025 production unit, serial number 337QXN, that has a manufacture date of 5/14/2025, It is located further from the wi-fi hub than the other unit and appears to be operating normally – although it is still in training mode.

One of my older units, a 2023 production unit (serial number 3344EK) was moved to a new location on August 24. Messages state that training can take up to two weeks. It is now well past two weeks and it is still in training mode.

Marsha

PS My oldest unit, a 2023 beta unit (serial number 333HGC) is working well. It is currently in a location quite close to the wi-fi hub. But over the last two+ years, it has been in multiple locations, and worked well in each location.

19

Hi Marsha,

If we want to get around the weak wifi issue. We do not need to wait 2 weeks for the training in the first location with stronger wifi. You can setup the device to the point where it is “In Training” initially. Then, use the “Move my device” function in the app and move the device to the location with weaker wifi. Although this is still not desirable since the device will often complain and will drain battery quite quickly if the final location has spotty connection.

For your information, the training of the device is not related to wifi connection. The app will say something about “device is offline” or “device is slow to update” if our backend is not receiving enough data regardless of training status. Training time is determined entirely by the quality of the installation and the soil profile

It maybe helpful if you could move 334RPU back to your original location and move another one of your good devices close to that location (but still at least 1 feet away from 334RPU to make sure their wifi signals do not interfere with each other) and do a side by side comparison so we can observe on our end.

We already have one RMA with another user where the device wifi is unusually weak. If it is confirmed to be the same issue we will send you a replacement.

Sorry for the belated response and thank you for your patience. We are still working around the clock to push out the irrigation recommendation feature.

Al

Engineering

20

Al:

I already had a comparison of two units, both being 2025 production units. One is the questionable 334RPU, the other is 337QXN.

Both units were installed in my backyard garden of small fruits and veggies. Unit 334RPU was closer to our wi-fi hub than 337QXN.

See the attached diagram.

Unit 334RPU location_09162025
Unit 334RPU location_091620251014×929 52.3 KB