Too Hot works on IFTTT, but Too Cold and Return to normal stopped

[CKurtis]

As of 3am 10/27 all the too cold and return to normal

triggers on IFTTT are no longer working.

I can recreate, I've de authorized, and re authorized

I've created new applets, HOT triggers each time, 

the rest don't.

My HVAC is in disarray.  Woke up to a freezing house.

Any Help?  

[Zhiheng Cao]

Nothing has changed regarding IFTTT trigger recently (however we did change IFTTT actions such as arm/disarm/enable monitoring, now you can select "all tags under the current tag manager', and we added update action, and 2 new queries to get current temperature and tag manager status)  

I would start from checking if the too cold and returned to normal events are recorded in the Event History (my.wirelesstag.net/eth/events.html) or not.  If somehow temperature monitoring is disabled after too hot it will not log back to normal or too cold event. 

--
You received this message because you are subscribed to the Google Groups "Wireless Sensor Tags" group.
To unsubscribe from this group and stop receiving emails from it, send an email to wireless-sensor-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/wireless-sensor-tags/69e0555a-81bf-442b-b48f-331e467a188cn%40googlegroups.com.

[CKurtis]

Been a long weekend of trying random things to rectify this with no direction.  Do the too hot messages go out differently from a port standpoint?  Like could this be a firewall issue?  I'm grasping at straws because I can recreate this issue over and over, from different tag managers and different tags, to different ifttt applets old and new.

Was all working great for years.  Could the multiple managers be conflicting in some way.  At one point in sniffing things out I put a new tag on my second manager and the new cold alert worked, and a flood of about 50 backedup messages came flooding out of the main tag manager... freaked my hvac out with multiple on off messages in a period of 45 seconds.

I have about 17 tags on one manager, and about 4 on the other.  Would love some input specific to the problem.  I already contacted IFTTT and they aren't having any issues.

I've had a few emails from support about this, but maybe its the weekend, but I wasn't getting a response.

Thanks

[Wayne Linton]

I have been experiencing a similar problem randomly since the end of July.  I don't use IFTTT, but the URL calls and phone notifications for "return to normal" do not get issued properly.  Sometimes they do, sometimes they don't.  I think it is the Tag Manager function to make these calls - to URL, to phone app, to IFTTT - and so I believe it is a fault in the Tag Manager.  The actual event "return to normal" is logged, but the triggering of the URL call and phone app notification is not issued.

I suspect it is a very similar issue to what you are having.

To address my problem, support suggested I use IFTTT.  But if that isn't being issued either, it is not a solution.

Wayne

To view this discussion on the web visit https://groups.google.com/d/msgid/wireless-sensor-tags/1336c8fb-dda7-49fe-881d-7c0ea4756acen%40googlegroups.com.

Virus-free. www.avast.com

[Kenneth Catron]

Hi, 

I also have this problem! The too cold trigger to IFTTT stopped working a couple days ago. I've tried resetting everything and cannot get it to work. Too hot still works. I still get the notification on my iphone that it is too cold, but IFTTT trigger doesn't seem to care. It has actually gotten worse yesterday. Initially when it stopped working a couple days ago, I'd do the "Check Now" in IFTTT and it would run the applet (multiple times weirdly). Now, it won't even run it. I only have one Tag Manager, and on Sensor. I have tried deleting the sensor from the app and readding it, as well as redoing all of my IFTTT applets, nothing has helped. 
I am glad to see I am not the only one having this issue!

Thanks,
Ken

[CKurtis]

Thank God I'm not the only one... I was checking each link of the chain from comcast, my router, cables, to the tags.  Sounds like it's "them instead of us".   Hopefully someone's looking into this.

[Kenneth Catron]

Update, 
Now my IFTTT ran if I manually clicked "Check Now" again, however it did the same thing as yesterday, it ran multiple times using old data:
I pressed "Check now" at 12:53pm and it seems to get like old data. 

On Monday, November 1, 2021 at 12:43:35 PM UTC-5 Kenneth Catron wrote:

[Kenneth Catron]

Looks like the last time my "Too Cold" triggered IFTTT successfully was 10/30/21 6:39pm CST

Hope this information helps, I am trying to think of some relevant data for you (@Wireless Sensor Tag Support)

All my events appear to be showing up in  my.wirelesstag.net/eth/events.html successfully. 

Thanks,
Ken

[Kenneth Catron]

I have now also tried revoking my IFTTT access from my WirelessTag account, then deleting all IFTTT applets involving WirelessTags and making sure everything between the two was disconnected. I then reenabled IFTTT access, and recreated the applets and the Too Cold is still not triggering. 
I then manually ran the applet using the "Check Now" in IFTTT and it did exactly what my previous post showed - It ran twice off of old triggers.

So something is not right and I think this issue needs addressing. It's either WirelessTag is not pushing the "Too Cold" notification to IFTTT properly, or IFTTT is not seeing it properly. Both Too Cold and Too Hot worked for me previously, only Too Hot works for me now. 

Thanks,

Ken

[Zhiheng Cao]

Thank you Ken for the detailed report. 

This pointed me to a subtle change in the server during development of the recently released 2 new query and the new update action. 

I reverted the change, and let me know if this problem is fixed. 

On Thu, Oct 28, 2021 at 12:30 PM CKurtis <ckur...@gmail.com> wrote:

--

[Brian Powell]

Not sure if it's related or helpful, but I decided to look through my KumoApp logs, and have noticed a lot more errors than usual beginning on October 21.

My KumoApp uses IFTTT to talk to my Wemo switch.  (When the temperature is too high, it turns on the switch. When too low, it turns it off.)  It's not unusual to get a "502 Bad Gateway" or "500 Internal Server Error" once or twice a week.  I saw a couple on Oct. 18, one each on the 19th and 20th, then four on the 21st, then 27 on the 22nd.  It's been about 20+ per day since then.

To be clear, sometimes the Wemo commands get through, but the errors are definitely more frequent than a couple of weeks ago.

In the past, I believe this has sometimes been caused by problems (such as memory issues) on the WirelessTag servers involved.

I'll keep an eye on it and see if the server change reverted today helps with this issue, too.

As an aside, I also have a backup mechanism using local URL calling to a Raspberry Pi, which can also turn on/off the Wemo from the local network without involving IFTTT.

Brian

[Kenneth Catron]

Nice! Is there anything I need to reset or get an update for?

You received this message because you are subscribed to a topic in the Google Groups "Wireless Sensor Tags" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/wireless-sensor-tags/0sg8V1B_T74/unsubscribe.
To unsubscribe from this group and all its topics, send an email to wireless-sensor-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/wireless-sensor-tags/CAOM4i21E3f9-4vnb2%3DV4mYSZhpT%3De3SBb_OCSd8aW3WL8oQn5A%40mail.gmail.com.

[Zhiheng Cao]

I am not sure if the fix will work, I just reverted some changes related to IFTTT Realtime API.  I am contacting IFTTT Platform support to get an answer.  

[Kenneth Catron]

My "Too Cold" just triggered! It actually triggered 5 times, but I think it was just catching up or something. I'll check it again here shortly to see if it goes through again.

Thanks!

Ken

[Kenneth Catron]

Boom! I think we are back. Thanks for the quick responses, glad I could help. 

[Chris Kurtis]

Mine are back and running again… I thank you and my chilly house thanks you. 

Sent from my iPhone

On Nov 1, 2021, at 3:13 PM, Kenneth Catron <kwca...@gmail.com> wrote:



[steve]

I really like the new disam all tags under tag manger but I cannot slect which manager I have 3 manager and I only see the enrty twice but they do not name the managers.

[Brian Powell]

I'm continuing to see relatively frequent errors from my KumoApp, so I suspect it's an unrelated problem with either the KumoApp server or the IFTTT integration.

As I mentioned earlier, this isn't a huge problem for me since I have a Raspberry Pi alternate solution.  KumoApp/Wemo/IFTTT integration has been flaky for many months, so I know I can't count on it for anything critical.

Brian

[Wayne Linton]

I've had occasional missing return to normal URL calls that are supposed to hit my RPi node red routine and terminate an action.  Most annoying.

Get BlueMail for Android

[Zhiheng Cao]

>    this isn't a huge problem for me since I have a Raspberry Pi alternate solution.  KumoApp/Wemo/IFTTT integration has been flaky for many months, so I know I can't count on it for anything critical.

Can you use  Raspberry Pi to trigger Wemo?  As far as I know Wemo does not have any open API and you can only trigger it through IFTTT,  or is there an open API I am not aware of? 

To view this discussion on the web visit https://groups.google.com/d/msgid/wireless-sensor-tags/f7d034e3-1915-4b2e-a366-11643fc4d0fen%40googlegroups.com.

[Brian Powell]

Another thing I do on URL calling to get a bit more redundancy is that I look for both too high/low messages as well as (every five minute) temperature updates.

The "too high" callback turns on the Wemo.  The "too low" callback turns off the Wemo.  The "temperature update" does its own computation and decides whether the Wemo should be turned on/off.

So I guess you could say it's triple redundant when combined with the IFTTT approach.

It's still dependent on the temperature sensors and tag manager working, but in general, this approach is relatively reliable.  Of course, I can tolerate a 5-minute control loop rate.

Regarding: "Can you use  Raspberry Pi to trigger Wemo?"

Yes, you can.  I'm using https://github.com/snagle/wemo, but there are probably other options.  It uses curl to send SOAP web service calls to a particular port.

Brian

[Zhiheng Cao]

Hmm that is not a public API; how do you find the IP_ADDRESS and PORT?  

Usage: ./wemo_control toggle|scan|on|off|state|signal|name 'IP_ADDRESS:PORT'

To view this discussion on the web visit https://groups.google.com/d/msgid/wireless-sensor-tags/4579086e-2430-4ccf-bd88-dd1a3915ca58n%40googlegroups.com.

[Brian Powell]

You use "./wemo_control scan", but you'll probably have to edit the script to suit your local network.

(E.g., my network is in the 192.168.1.2 to 1.199 range, and I only "expect" one Wemo device.)

And all "scan" really does is query one of the three likely ports (49152, 49153, and 49154) for /setup.xml to see if it looks like a Belkin XML file.  You may also want to increase the timeout by a little bit, too.

If you can't figure it out, send me a direct message and we can work on it.

Brian

[Brian Powell]

I should add that I also looked at https://github.com/iancmcc/ouimeaux

It seems more robust, and more complex.  I think I probably went for the simplicity of the other script, but I don't really remember why I chose it.

Brian

[CKurtis]

As of 8pm CST yesterday, the issue is reoccurring.  woke up at 2am very cold and concerned for my kids.

Again it seems the too hot messages are getting thru to IFTTT, but not the others.  PLEASE HELP?!

[Wayne Linton]

 In recent months I have been experiencing missing triggered calls for "return to normal" - phone app notification, email, and URL call  (I don't use IFTTT, but I suspect the "return to normal" function would trigger them all, or nothing.)  Doesn't always miss the triggering, but frequently enough to be troublesome.

Since changing the battery in the tag, this has no longer occurred.  I don't understand why, because the "return to normal" is logged in the tag manager has having occured - it just doesn't trigger the requested functions. 

So just as a check, maybe put a new battery in the tag and see if the "return to normal" IFTTT function is triggered.

Wayne

To view this discussion on the web visit https://groups.google.com/d/msgid/wireless-sensor-tags/b22f70e5-0552-442c-a1b4-afe94d0705e0n%40googlegroups.com.

Virus-free. www.avast.com

[Chris Kurtis]

This was an issue about a month ago, they had changed something that caused it. I’m thinking it changed again. I have about 20 tags. Would be a helluva coincidence for them all to have a bad battery.  All of the too hots show, none of the others. 

Sent from my iPhone

On Nov 30, 2021, at 10:11 AM, Wayne Linton <wayne.l...@gmail.com> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/wireless-sensor-tags/40740344-2dfc-2a9f-4f0a-4426a598c68e%40gmail.com.