I got into Z-Wave-based HA around 2010. Back then there ware virtually no LED light bulbs (or they were insanely expensive). Within the past two years, I gradually switched to LED’s and replaced most of the old-generation dimmers that didn’t support LED’s. So the 5-year upgrade cycle’s been working for me so far.
Controller is a different story of course. I started with Z-Wave remotes in 2010, then upgraded to Vera Lite in 2012, then switched to ST in 2014 and now ready to switch again. So yeah, controller upgrade cycle is close to 2 years.
I hope you are right. A two year cycle, or even shorter, sounds really good about now. Getting off my current “solution” and onto something else feels right.
A hobby implies that you draw pleasure from it… With ST I most certainly do not anymore. There’s always been a carrot at the end of the stick that is the perfect smart home. But after lots of waiting and rhetoric from ST that alludes that in the near future this will be possible, I have definately come to the conclusion that it’s definately all stick
If you offer to take responsibility away from a user for the maintenance of their data and the fundamental control of their home automation then you undertake a responsibility to deliver a service that is as near as possible to 100% reliable.
When you don’t provide some other mechanism for local control or backup/restore for each user you are just courting disaster.
When you FORCE people to use such a service then you absolutely must deliver - and as repeatedly demonstrated you’re just not doing this.
I personally believe the cloud architecture is fundamentally flawed for use as a core mechanism for home control. These incidents just reinforce that belief. Every action including logic and schedules should be implemented locally with event reporting to the cloud as a secondary concern. It must just work, as expected and every time. By all means supplement your system with more involved cloud based devices/services as you wish but accept their response time and even availability is not guaranteed.
Just consider that if each user has had to spend just an hour remedying the problems caused by this issue and the now loss of DB state what cost that is in cumulative $ man hours to remedy. Millions of $ possibly ? It’s many orders of magnitude more than employing just a few more testers at ST in advance or some engineers to at least implement more local control / backup. This was after all the much touted main differential feature of the V2 hub.
I know you say this issue only impacted a subset but a subset is really anything <=100%
It’s really looking like this dogged dependence, nay insistence on the cloud paradigm for even basic functionality will be the Achilles heel for ST. C’mon guys - you insist it’s the right way to implement this - deliver.
The problem here is that it’s our money not theirs, so as long as we don’t pay for a maintenance contract, then most likely it will continue to be our own pocket money spent on maintaining individual system. Secondly, like @Mike_Maxwell and others mentioned, these incidents affect mostly power users. I’ve read people saying that stock apps were affected, but I am yet to see proof of that.
Home Automation is our hobby, not SmartThings. ST is just a gadget. And when a gadget takes away the fun, then you move on to other gadgets that bring the enjoyment back.
I can say for certain, the last db corruption round left me with several SL rules that were missing configuration data. This, perhaps, led to months of ghosts that had rules firing at seemingly random times, etc.
I don’t think it’s particularly important if this round effected stock apps or not. It can, it has, and until there is a proper mitigation - it will.
That puts this family of issues squarely in the realm of consumer users. I am not sure we information to establish it only effects mainly power users. Our pool of users here, in this thread for example, are what I would call power users and that echo chamber may make it seem that way.
I don’t want to be the @bridaus of this thread. I dodged the bullet this time around so I won’t go further to jinx it, although I may just have done that. lol
Looks like the corruption has killed my Netatmo connect as well. Unfortunately as I’m in the UK ST aren’t interested in providing any support as it US only. So if anyone knows of a fix on that I would be interested or if anyone got a fix from US support I’d be interested.
I am not sure if the Netatmo issue is related to the issues earlier this week. This might be coincidental (it also might just be affecting Brians), but it has definitely hit our radar. If you haven’t yet - please send a note to support so we can flag and trend the issue.
I haven’t seen an official answer to this. Is it time to rebuild or am I still holding out hope? Can we get an official YES/NO? Lay it on me, I can take it.
ok I have rebuilt all 100+ rules into Core and now ST will not change SHM state properly, even manually, routines fire several minutes later and all other kinds of odd things keep happening. When will ST actually fix this meltdown and not just make the marketing page look good???
bamarayne
(Jason "The Enabler" as deemed so by @Smart)
240
So I took all this as a clear warning and have now removed RM from my apps i’m glad to say I’m finding CORE very functional and not a big learning curve at least for basic rules. So far everything is working looks like a good move.
I’m now just worried about the rest of the community supported apps some of which I’ve had to reset and some of which I’m not sure if they are ok or not yet. It’s going to take a while to check them all.
