I’m afraid that I can’t help right now either. The last time my system watered was Sept. 23. After being in a severe drought, we suddenly got rain (and this weekend WAY more than we want!!!)
No worries, if it happens again I will followup with IFTTT.
I realize this thread is quite stale but I am experiencing exactly this problem.
If I water my grass with Smart Cycle I get inaccurate and duplicated entries in the Google spreadsheet produced by the Rachio service in IFTTT. If I turn Smart Cycle off the entries are good.
Is there still someone in charge of the IFTTT interface?
If so I would be happy to help reproduce the problem.
I’m still experiencing the same problem also.
Seems like we are having issues with IFTTT all around, I am trying to do some testing but having trouble replicating the inconsistency. I am still trying to test and can talk to our development team to further investigate!
Not sure if this will be of help, but as a workaround this (link) may be a way to get rid of duplicates.
If opening the file with Microsoft Excel, there is a similar function called “remove duplicate values” (link)
Gene
1 Like
Thank you for trying to replicate this problem.
Here is a bit more detail in case it helps.
I have 5 zones for grass. I am using a single Daily Flex schedule to water them (which is great by the way!). When I used Smart Cycle the system interrupted each zone 2-4 times with soak periods (also great). But under those conditions the IFTTT service posted twice. Once at the end of the 1st cycle and then it made a duplicate entry following the last cycle. It also inexplicably posts durations which are from 3-7 minutes more (?) than I get by adding up the actual zone rune times.
When I turned Smart Cycle off both of these problems went away.
This particular Rachio IFTTT Applet (ID = 54997486d) is supposed to be posting “when watering stops”. Therefore it seems to me that the 1st post is a bug. Watering has not stopped/ended yet following the 1st cycle.
I will turn Smart Cycle back on now and post results here when next my grass is watered.
1 Like
Here is a theory which may explain these double posts AND why the problem is apparently hard to reproduce.
According to IFTTT documentation it is normal for the Rachio service to be polled about every 15 minutes. See https://platform.ifttt.com/docs/faq
The service is supposed to return a list of 50 time-ordered events each distinguished by a unique id. IFTTT is supposed to only trigger on NEW events. But here you show the same event id with two different timestamps.
I suggest that creates an ambiguity. Logically an event can only occur at a single point in time.
If the duplicates occur within a rolling 50 event window as buffered by IFTTT might trigger once. If the duplicates are further apart it might trigger twice.
Hope this helps.
1 Like
That FAQ article is helpful. It notes that:
“When you implement your trigger endpoint you send us a meta.id value with a unique ID for each event. IFTTT will keep track of which IDs it has seen in the past and only trigger for new IDs it hasn’t seen before.”
My guess is that since your watering has technically stopped due to the smart cycle, the system is reporting a stop with a new meta.id value. I think it only really occurs with the first cycling due to the 50 time ordered events polling method, as many other events can be occurring between schedules.
I appreciate the pragmatic suggestion. For me it is more satisfying to get the underlying problems fixed than to work around them. To each his/her own.
Update: I turned Smart Cycle back on and indeed the issues returned. As you can see in the screen shot below I had double posting and duration discrepancies on June 7. Turned Smart Cycle off and problems went away. Turned Smart Cycle back on June 21. Problems returned.
I see elsewhere it has been stated that the reported duration inconsistency is because “rounding”. IMHO it would be less trouble for Rachio if the duration numbers were made consistent rather than explaining “rounding” over and over again to customers. Just sayin.
Hi mckynzee,
Can you please provide status on this? Has Rachio been able to replicate the double posting? Does Rachio feel it’s a defect with IFTTT and therefore insoluble?
Thanks
@piperspace So far we have been able to replicate the issue through building similar schedules in IFTTT. The issue seems to be related with the way IFTTT “polls” the events from Rachio’s servers. At this time I can not think of a fix off hand, as I am no developer, but the only way I can see this being fixed on our end would be some sort of way to edit the way the meta.ID values are being generated, as this seems to be what IFTTT uses to differentiate one log from another.
So long story short, maybe we can do something about it, but I can’t be sure. I would recommend sending a support request to IFTTT as well and make some noise about this. Hopefully there is something simple they can do on their end!
Thanks for confirming you can reproduce the problem. That is helpful.
I am a commercial developer. It is clear to me there is a bug in IFTTT service code authored by Rachio. When Smart Cycle is on - your code is generating a bogus event after the 1st cycle. You should stop doing that.
I get it that your programming team may have other fires to fight just now. This is a reporting issue not a problem in basic function. However, best practice for a mature software company would be to assign the defect a number and severity code and put it on a list for attention by programming in future. Can you do that? Otherwise it kinda looks like you are stonewalling the issue and failing to own up to a known defect in your product. Thanks.
Hi @piperspace,
Thank you for the feedback. I can understand your frustration with the experience as we investigate this issue. Please know that we are in no way trying to displace blame or “stonewall” our users. Thanks to the details you have provided, along with @kpontis’ internal testing, we’ve replicated the issue on our end and have in fact logged it for our development team.
Please understand that our developers are hard at work developing new features and improvements for future updates. We are working on prioritizing defects as they are discovered, and we have only just recently been able to replicate this issue on our end. Rest assured that we are aware of the issue and are working on prioritizing it, however in the meantime it may be beneficial for IFTTT to offer their evaluation on the issue as well (just incase they are able to adjust anything on their end).
Thank you again for your continued patience while we work on reaching a resolution!
1 Like
Very well. I have posted a link to this discussion with IFTTT support.