I know this is nothing NEW here, but it IS news where my system is concerned. So, just FYI…
Even though I’m still in ‘hands-off mode’, I keep seeing new things broken day after day;
things that weren’t broken the day before.
Rules not firing upon obvious triggers (like sunrise/sunset, or Mode Change, etc),
one rule sent its change data to the wrong device (what?!?),
upon clicking a light switch that controls two smart bulbs, only one of them comes on, or maybe the other one, or none at all,
BloomSky integration is completely hosed
the device handler and smartapp by @slagle is not updating, EVER unless I do it manually,
AND the WU-PWS smartapp (i.e. Weather Underground PWS Connect) is no longer sending its weather data to my Weather Underground account…
and on and on…
p.s. Don’t worry, I’m not freaking out or anything (YET), and I’m not expecting any of this to get any better yet, because I know the PROBLEM hasn’t even been identified yet, let alone fixed.
Hi - please let us know if you have problem tickets in with support. We cannot track down specific issues without that. If you would like us to look into these things on your account, let us know. It is especially valuable to be able to compare account setups because many times there is a SmartApp or rule collision that causes apparent platform issues but without multiple data points it is hard to diagnose.
It’s a little hard to know when to submit a new ticket. I submitted one last week when my SHM was inaccessible, which resolved, but then I had a good night routine that went rogue and started running several times a day. Then yesterday one of my cree connected bulbs stopped talking to ST.
What’s due to day-to-day unreliability? What’s due to this major platform disruption that, to this day, remains unexplained and unclear to what degree it’s resolved?
Update from yesterday hasn’t changed, we are testing many improvement efforts on the platform. Right now we are sensitive about releasing them without being sure they won’t do negative damage.
I will be sharing the improvements efforts we push live as we do them so you all can stay informed.
For me the issues I noticed began Sunday Night/Monday Morning. The single device I have stopped sending notifications. Monday night 3/21, I tried using the app to open the GD, no worky. Given the hints of DB issues, I figured by last night it was safe to try a remove and re add. (exclude/include). Since then it’s working better than ever (since I got the hub V2 and IRIS gdo a month ago) Today for the first time I received a notification when not connected to my home wifi.
I am new to this and have several suggestions to improve this community format. Venturing off topic here a bit, what this community needs is sticky tutorials. Right now I find it incredibly time consuming to navigate through hundreds of posts to cobble together a single solution, as a critical update can be buried 300 posts down. Either the Original Post needs to be updated upon code modification, or the thread needs to be closed, archived, and then a new one started with relevant current info on the top. This is off topic enough. People here have wonderful insight, as someone new to coding, I just find it hard to track all the info. Maybe the mods just need to get better at killing duplicate threads.
Back to topic a little - One thing is for sure from my experience thus far, this platform is still very much a beta experience despite it’s age. I love the fact that it’s flexible, and that the users get to tailor make just about everything. Reminds me of myspace, you could put your own html code to make your page just the way you wanted it within the sandboxed environment. This does however keep the platform in a constant state of beta as seemingly minor updates break things that are not immediately visible. This platform’s quest to make complex systems, user friendly, is sort of shooting itself in the foot…Do you even remember myspace? We are pretending unsecure and uncontrollable cloud infrastructure is wonderful. Has anybody thought what happens if ST get’s infected with ransomware? With as flexible as this system is, even an hour old backup is going to cause a fair amount of users to have to redo something. From what I have learned so far, The Android SmartThings APP doesn’t know the database entry for the device/devices is gone, leaving users to guess on how much they have to redo. Fortunately for me, it was just one device. I can see now though, I am better off using a Zwave stick. If I have to mangle code to make something work the way I want it, I might as well have full control of the dependencies.
I am in this boat. Everything was working within normal degrees of wonkiness until then. Now SHM is completely unusable and I have contact sensors showing the wrong state even though the activity log seems to indicate that they should be closed. I’ve been on Ticker since last week, so I don’t feel like that is the issue. Schedules have been good, just not the stuff that the schedules fire off. Just going to ride this out and hope for the best…
One thing that might help you disarm is to change the arm status from Arm/Away to Arm/Stay and then to Disarm. It worked for me when I was having SHM issues.
SHM started not working again today. Cannot get it to disarm. This is crazy…it has been over a week. My guess is that the performance improvements were a result of people on installing the SHM smart app. Now that everyone has reactivated it, we are all having the same problem again. Ughhh!!!
So you started to have SHM problems again as well today?
mdawson
(Different Computers. So happy with Indigo.)
455
you know what’s weird? For all the problems I’ve had, SHM has never been a problem.
Though I have to say, it’s hard for me to notice the sorts of problems people mention here because I can’t really recall a time when Things Just Worked Smartly to the point I felt like I could rely on them.
How can you implement a new scheduler that is supposed to definitely work better when you don’t even know why SHM is not working properly right now? Is this new scheduler going to make SHM work and arm/disarm like it supposed to? If not, why are you not releasing a new SHM app that actually works? I have tried to be patient with all the problems this last week, but it honestly seems like ST doesn’t have their priorities straight. Who cares about a new scheduler when you can’t even get your security system to disarm! I am all about new features and possibilities of the ST platform, but I would rather see the existing feature set work properly first. If you start adding to and changing the platform before fixing the current one how will you ever be able to diagnose a root cause and fix it.
I pulled my screaming banshee (siren) out of my SHM configuration last week. Things have been quiet while I have watched the platform twist and turn through this thread. I was largely unaffected (60+ devices) until my banshee started screaming this evening at dinner. No idea what set it off. No smartapps are linked to the device. I took the hit, told my wife it was me, and apologised. She bought it.
Ah well, hopefully it all works again soon.