UPDATE: Recent SmartThings User Experience & Platform Performance

I am new (purchased hub about 3 weeks ago) so came into this at a bad time it seems. However, my problems have been limited to difficulty adding rooms and devices due to the your not authorized error or the check your network error.
The problem I have now is my Wemo kitchen light switch works with the ST app, Echo, timers, routines, STiles, but will not automatically update its state. I have to refresh it in the ST app to get it corrected. Renaming didnā€™t work either. I should mention that it was working until all this hell broke loose.
I donā€™t use SHM. I only have the wemo, ecolink motion, garage door opener, cree bulb, leviton switch, and GE 3 way.

Iā€™m not sure if this is the same issue I had, but some devices donā€™t do a great job at updating. Thatā€™s not really a ST system issue as much as it is individual devices. My CREE bulbs donā€™t update either and will show the incorrect level until I do a manual refresh. Really a pain, but once you figure out whatā€™s happening the frustration level drops.

If my hub starts dispensing Blantonā€™s Iā€™m going to be the most forgiving man on the planet.

4 Likes

Wouldnā€™t it be great if it just turned out to be a specific app? Unfortunately, Itā€™s a musical chairs problem. This kind of thing has been going on for at least a year. It has nothing to do with what you were running or support wouldā€™ve already told us.

Sometimes it has to do with where you live, but not usually. most of the time itā€™s just a matter of luck. ā€œWhere were you when the music stopped?ā€ Kind of luck. Some people have no problems at all because they were in a good place and there was an empty chair right next to them. But on the next round they might not be so lucky.

If you go back and read the technical explanation from the SmartThings staff way up thread, youā€™ll see that the problems are just caused by ā€œpressure on the databaseā€ Not by anything to do with specific apps or specific devices or specific accounts.

There are a lot of very smart, very technical people in the community and they have spent hundreds of hours over the last year or so trying to find common factors that we as individual customers might have some control over, and ultimately there arenā€™t any. The common factors are that the data base gets overloaded and some unlucky person gets zapped.

This was most obvious last fall when one member reported that there was a device in his account called ā€œdining room sensorā€ ā€“ ā€“ and he didnā€™t even have a dining room!

Another person carefully explained that he never had any problems because of the 17 steps he had done to make sure that everything on his account ran perfectly (the rest of us threw virtual tomatoes at him, of course) and then a few days later he started having problems as well. :stuck_out_tongue_winking_eye:

If you feel like you want to collect random account data, please start a new topic for it so we donā€™t get flooded with dozens of posts about whether people have a specific smartapp or not. But the answer still going to be just luck. Thatā€™s how database flooding works.

The Following thread from last October gives a lot of previous examples

Time based events failing?

3 Likes

Yea this is the first time since I got the system that Iā€™ve started actively looking for replacementsā€¦ Competition isnā€™t as feature rich, but Iā€™m sure the features that are there workā€¦

1 Like

Depends on the competition. And the features. :wink:

Every system has pluses and minuses. In general, the ones with more reliability in the same price class have fewer features. (Indigo, for example, is rocksolid, and runs locally. But it only supports zwave and Insteon, and even then no zwave door locks. ( it doesnā€™t support the security command class) ) So it just comes down to what you specifically need as to what a good match will be. Detailed discussion here:

Haha, that would be me - good times! Whenever things go bad with ST, I pull up that screenshot to remind myself that things could be worse.

6 Likes

I will say thisā€¦ Iā€™ve been really frustrated with ST lately, but this weekend was the first time in probably 4 months that everything in my system worked properly, including scheduled timers and the gradual dimming functions that I use ā€œGentle Wake Upā€ for.

Of course, now just starts the countdown for the next time Iā€™ll have issues though. :smiley:

2 Likes

Yea Iā€™m messing with the Indigo demo on my mac ā€œserver.ā€ The locks is a big deal breaker. I see it has a lot of potential, it just looks a little behind the IOT ball.

I know itā€™s completely unrelated, but I thought it was funny when I checked the status page sometime yesterday I think and got thisā€¦

4 Likes

Ah, those were the good old days.

Iā€™ve not been here as long as JD, and I was one of the ā€œgive them a breakā€ crewā€¦up until my 6th alarm going off because the SHM didnā€™t turn off on time. Iā€™ll still give them a break, but I wonā€™t recommend ST to others, at this time.

Most of us arenā€™t ready to abandon this ship. I still think the Hub is a useful device, I still like it. Butā€¦

In the end, we, the users, canā€™t do anything. The ST engineers should have logging and measurement tools in place to be able to determine where the problems exist. Frankly, I suspect the whole architecture may be flawed, if theyā€™re still trying to determine the root cause after months.

What irritates me, though, is this make-work the ST people have us do. If the issue is the back end, then why does using a custom rule work, but not SHM? It probably doesnā€™t. Or if it does, itā€™s temporary. I suspect the whole reason weā€™re told to make change is to give us something to do, so we stop complaining, while they throw equipment at the app.

But these changes could be starting to muck our systems up, with their convoluted equivalent to ā€œfollow these 17 stepsā€.

(This weekend was the first time I heard the back end was using Cassandra. Cassandra and Java. Ick. Sorry, Java bias.)

JDā€™s point is good: this really isnā€™t the way to find the problem. The ST people will have to find the problem.

2 Likes

My concern is this: It appears that this may have started out as an overload or attempted change to the scheduler, then this eventually resulted in database corruption - weā€™ve seen many cases that point to this. If thatā€™s the case - no matter how it happened in the first place, anything short of a restore from backup on their side (very messy), the data (devices and rules, etc.) wonā€™t be coming back on its own. Intervention on our part - such as rule rewriting/replacing or changing device names to force a data rewrite, etc., may be the only way to clean data. I also fear that the existing corruption of data on their end is continuing to corrupt more data - and create new issues. If we truly have data corruption, we have a very real and ugly mess that may self-perpetuateā€¦

knocks on wood Iā€™m at about 36 hours with no issues. Still not confident enough to install a siren, but even SHM has been behaving. Which is nice, because we have a 1 month old and I have SHM set to disarm and arm based on our master bedroom door open/close, which happens about 3x a night when going downstairs to fetch bottle parts.

I agree, a corrupted database is probably a worst case scenario.

Iā€™m just not going to require close to a hundred rules over three apps, only to turn around and do it 10 more times.

My plan is to wait it out, perform maintenance, and when weā€™re running again Iā€™m taking the entire system down and starting it from scratch.

1 Like

That makes sense. My housemate just announced this morning that I have to choose between SmartThings and breakfast. :flushed:

Since the last iOS mobile app update broke voice navigation, he has to do everything with the app for me. Usually in the mornings he makes breakfast, puts away dishes, and checks the laundry before he goes to work, but he can sort of do all those things on autopilot.

Since the DST debacle heā€™s been spending about 10 minutes every morning checking things for me in the app, and I think heā€™s just tired of it.

We may just unplug our hub until things stabilize and the mobile app is fixed. Because, you know, breakfastā€¦ You have to keep your priorities straight. :stuck_out_tongue_winking_eye:

5 Likes

Thanks for all the updates guys and gals.

Weā€™re pushing a few things throughout the day today we are confident will continue to help. Weā€™ll keep you posted.

3 Likes

Youā€™re still SOL?

Cā€™mon, guys. Wake the :speak_no_evil: up.

1 Like
1 Like

To be fair, they offered to try and find me the previous build for the mobile app, but then Iā€™d be out of sync with the rest of the world and it just seemed like too much trouble. Whatā€™s the point in being able to take screenshots if the screenshots themselves are useless to anyone but me? Plus that would be a fun conversation to have with support: ā€œno, I canā€™t update the mobile app to the most recent version because that one doesnā€™t have voice access.ā€ Weā€™d both go way off script. :wink:

Iā€™m hoping theyā€™ll have the mobile app accessibility issue fixed in the next update. @david.mccrindle indicated he felt it was a pretty high priority.

5 Likes