You're slowing down in your old age.. I would have thought you could have done it in 38!
2 Likes
Back to the topic of slowdown solutions. My HE is back to its speedy self after removing 6 peanut plugs from my mesh. I tried with power reporting enabled and disabled, with the generic zigbee outlet driver and the Peanut driver, and saw no difference. I wasn't really seeing dramatic slowdowns, but consistently saw an increase of about 500ms in my virtual switch in Hub Watchdog and rebooting did not make any difference to this time. This is where I'm at now:
I also think that incomplete removal of devices from rules and apps may also be causing some of the issues I have had in the past. I installed 4 Sengled color bulbs last week. I removed 4 Sylvania RGBW bulbs from a group and added the 4 Sengleds. One of the Sylvania bulbs still turned on and off with the group device it was removed from. I had replaced a bulb that was in a lamp in the same room with one of the Sylvanias or I probably wouldn't have figured this out for a while. I had to delete and recreate the group to get this to stop happening. I also had some strange errors being thrown on my main hub whenever my hue motion sensor (which is on a different hub) was triggered active or inactive. This device wasn't even being mirrored to the main hub and didn't have anything on the main hub associated with it, but had at one time. The only way I was able to stop the error was by deleting the device from the remote hub and rediscovering it from scratch.
BTW, I had already tried a soft reset on both hubs when I was trying to figure out what was causing the error, and removed all remote hub devices to let them be rebuilt and the error continued to occur.
@bravenel , any idea what would cause these issues. They seem to be related.
The one thing that helped my hub was NOT using all devices on every dashboard. I think they even imply that in one of the video chats, but unless I missed it I don't see that warning in the documentation or in the app itself.
Even though that setting is available, I believe it to be a bad thing to have available. It would be nice to have other ways to sort/select items available to the dashboard, but if that isn't in the works, I would suggest the Hubitat team just remove that setting.
1 Like
I'd disagree for one reason: new users. New users with only a few devices setup need it to be as easy as possible to initially configure a dashboard.
Wow, I just tried this, and anecdotally, it has made a marked difference.
I will consider moving the dashboards off my main hub.
1 Like
Is this regarding auto-filling a dashboard, or the toggle setting in each dashboard that says 'use all my devices'?
In the Dashboard App:
Choose to allow access to all your devices to this dashboard or pick individual devices. This does not automatically add the devices to the dashboard, just makes them available to be added.
Use all your devices
This setting.
1 Like
If you only have a few devices, it should be simple to select a couple devices to use on dashboard, and give a more detailed explanation in the app, or in documentation why/how to select the devices they need.
It should be easier for a new user to pick their few devices than someone like me who has to sift through approximately 150 devices.
Iām with you on this. When I first changed over everything to Hubitat, I had over 100 devices and didnāt know what I was doing when setting up the dashboards. The first one I did was the automatic setup and when the dashboard came up I said some things that I canāt post. I couldnāt even see the names of the devices, so I deleted it, and decided that picking what devices I wanted to include was the way to go.
I just did that on one of my hubs. Immediately seems faster. No real data to back it up, yet, but it definitely feels more responsive.
1 Like
This is not what I am seeing. I turned on one of my 10 Iris V2 plugs for 10 minutes. And this what I saw.
After the first minute, where power is stabilizing, I do not see another report until I turn the device off at the 10 minute mark. Not every 10-30 seconds.
Also, during this 10 minute period, I received two entries for other Iris plugs, both for cluser 0x0019 which is the OTA cluster, not the power cluster. So, I did not receive ANY messages about power reporting from any of the other plugs I have connected to the system. I am using the Generic Zigbee Switch driver that it paired with when I added them to the system.
If you are seeing reports every 10-30 seconds, I'd be interested to hear what granularity you have your plugs set to. I have mine set to 5w for this device because I do not need granularity down to 1w. I would suspect that would greatly increase the number of messages.
2 Likes
I changed the refresh time for local and cloud dashboards. I feel that if I'm not home, I don't need an instant refresh for most things. I also don't share many of my dashboards to the cloud since I use Homebridge for most of my dashboard needs.
I don't share everything with Alexa and most of what I share are group devices. I have seen a lot of posts on here about issues with the Alexa integration and wonder if it related. I have never had an issue with Alexa integration.
If it's going to be documented, just provide a warning in the app/documentation about performance being slower when you include a large number of devices.
I just spent time last week reorganizing my dashboards and deselecting "Use all your devices" to speed them up. You're still going to have to sift through the devices that you need to select for each dashboard.
Many thanks for your driver. Ever since I changed my Samsung (2018) plug to the Iris Driver, the Better Laundry Monitor app has worked flawlessly. 
1 Like
A few questions related to slowdowns
-
I have a early SmartThings presence Fob (using the V2 driver). It seems to send a Zigbee event every 20 seconds and is the chattiest Zigbee device on my mesh. How much of a problem is this?
-
I have Node-RED logging set up, including polling the Maker API every 15 minutes for presence and motion. I feel I started seeing slowdowns soon after I enabled that but I also made several other changes in the same time frame. Has anyone had experience with MakerAPI causing slowdowns? The all details query is pretty slow.
-
I have an app I am working on with maybe 10 schedules and a decent amount of state (one big array). It only runs about every 5 minutes, but the schedules are randomized. State changes happen every 30 minutes. I would think the C5 could handle this pretty easily, but I don't know anything about the scheduler or state database implementations.
Thanks
Thatās how the device is intended to work. I donāt believe that, by itself, it will cause a problem for you.
1 Like
I'd like to ask everyone who reads this note, what they think about this suggestion.
Perhaps this has already been thought about, perhaps not.
Perhaps we should not have to do this, but perhaps it will help many users.
We can make this optional, by allowing the user to opt-out, by putting something in "settings"
Why not make, as part of the daily backup procedure at 3am, a period of "maintenance". As part of that "maintenance" the hub is rebooted, or even some "soft reset procedure" is part of this "maintenance" period.
P.S. A number of other competitors do this daily (e.g. Vera, etc.).
Please consider this a "feature change".
By making it optional, you can ensure that everyone who doesn't want to occur, will not have it occur.
There may be a small number of users who will not be able to do a reboot - but they have will issues anyway.
I know that this shouldn't have to be done, but isn't this an expedient way to get rid of the slowdown issue?
1 Like
Definitely make it optional. This is a Band-Aid solution and it will result in lost events. It's the sort of thing SmartThings did and the reason I left them. The only time I reboot is when a new release comes out.
1 Like
Same here. I probably don't even have to do that, but it is just a habit from doing updates on other devices.
2 Likes
