Reliability is gone

I'd also like to point out that this issue (slowdown, reliability, etc.) is the other side of the coin called "local".
Other companies devices can run HUGE streams in the cloud, and it may not affect what goes on locally. However,because the HE runs locally (almost exclusively), it doesn't have that ability to offload processing to a cloud.
I like the local, but it means that we have to be MUCH more careful what we run. We can easily "blow it out of the water".

1 Like

I tried disabling the rule triggered by my washer power meter and my issue might actually be solved. More test runs needed to be sure though.

Interesting, I have notifications by the washer and the dryer, 2 HEMs in the same hub, no issues, but I use Message Central instead RM...

I also don't have any noticeable issues any more. I have a theory, but someone will tell me I am wrong.

I uninstalled IFTTT. I tried disabling a bunch of user apps including Echo Speaks. I didn't find any one app that made any difference in slowdowns.

What really worked was messing with dashboards. I had every device available on every one of my 8 dashboards. It was bringing the hub to a crawl. I now only have the devices I need enabled per dash. It made a huge difference in speed of loading the dashes, and general GUI speed.

3 Likes

My Iris outlets are set to report at either 1W or 5W. Anything higher and they donā€™t report when the final cycle is done: value stays at 6-7W when it finally stops, never reporting zero unless manually refreshed. The built in driver only allows for preset power thresholds instead of user specified and doesnā€™t give the option for a min/max time between reports which Zigbee spec should allow for. So at that reporting level, it can be triggering the rule up to 1-2 times per second. Might have to change the way I go about thisā€¦

That actually makes sense along the lines of resource exhaustion.

It would be great if there was some kind of profiling tool or something like ā€˜topā€™ that could show what is consuming resources. I hate that itā€™s a black box that I canā€™t even really debug on my own. Still, better than SmartThingsā€¦

I mentioned in a previous note, that it's important after rebooting the hub, to turn off the Hub for a few minutes, so the the zwave radio gets "rebooted". For a long time, I thought that the only way to do that was manually - turn the hub off, and then a few minutes later turn it back on.
In a separate note, on a separate matter @SmartHomePrimer stated that he found a way to disable the zwave radio. That means we don't have to turn off the Hubitat, just disable the zwave radio.
The following is a small app which shows how to do that in RM:

P.S. Post commands don't always work in the latest version 2.1.17.118

1 Like

With regards to turning off and rebooting the zwave radio... would this be important if using only Zigbee as well?

If you're using only Zigbee, and you want to put your Hub into a mode where all devices will start searching for a way to the hub, then you have to turn off the zigbee radio for approx. 20 minutes. Note: this may have reporcussions on battery life of some zigbee devices. The code for doing that is as follows:

I'm using sharptools and only have like 20 devices between three dashboards. I haven't edited it in a long time. It was working fine for months before all this started. I also deleted most of my apps including echo speaks even though they were all fine for months before this started. If it can't run the few I have left in the screenshot above then I don't want it. After the holidays I'm going to do a full reset and repair everything. If that doesn't work I'll be moving on

Support actually messaged me here back in September because that's when all this started. They asked if they could look deeper into the problems on the hub. I said yes and gave them my information and never heard back.

2 Likes

Other than bobby - is there any Other support staff ?
On twitter they said they were rounding out to 2000 users, you cant tell me bobby is the only support for 2000 users..

And if he does, maybe we should call him "Super"-bobby from now on :rofl:

1 Like

I believe one of the Hubitat guys, @bravenel I think, suggested this a while back . . .

:upside_down_face:

1 Like

If the 10:30 trigger is a backup for the 9:30 trigger it's not being evaluated the same in your IF-THEN statement. The 9:30 will execute the IF and the 10:30 will execute the ELSE.

I just wonder what the point in forcing Zigbee network to rebuild(panic) with every reboot? You are not moving devices around the house every day I would imagine...

Not with every reboot.

But perhaps, only once a week. (If you've done some moving around. If not don't do it.)

P.S. Other competitors (e.g. Vera did a Zwave Repair every night!)

Vera also reloaded the LUUP engine every 6 minutes. Didn't make it a good thing.

2 Likes

I don't think that this makes sense either. My last ZWave repair was done over two months ago and I have no problems. I have not added/removed/moved any Zwave devices therefore I don't need a repair.

A Zwave repair is really "only" needed if you

  • add a new Zwave Device
  • remove a Zwave Device
  • factory reset a device and rejoin it
  • physically move a Zwave devices from one location to another
3 Likes