Outside the the US it's way more than 2 days 
If you're sharing devices between hubs then you've defeated the entire point of trying to speed up response time and made it much worse. That is really cool though. I was not aware of that. So theoretically I can create a rule where if my parents smoke alarm goes off it sounds an alarm at my home so I know? That could be useful especially as they get older.
Yeah, I could understand having a backup hub for in case something happens, especially in places where it might be difficult to get a new one in a quick time frame. Somebody on here said something about having 5 hubs though. That's complete overkill and waste unless your home is 30,000 sq ft or something and at that point if you have $150MN for a house, you'd probably spend $50k to have a company come and set it up with stuff that makes our automation look like total crap.
1 Like
The Author of HubConnect has greater than 500 devices in his home. It's a crazy number for those of us that haven't hit 250 yet, but indeed giant installations occur.
There have been a lot of discussions about the 'value' of multiple hubs. I always begin such discussions by agreeing with you. For most people, a single hub is adequate. But that leaves a LOT of wiggle room for justifying multiple hubs. It's rather trivial to justify actually.
Cost is relatively modest. Therefore as the 'price of a solution to a real problem' makes the decision easy for most. I recognize we haven't had a sale in a while, but a hub is typically in the range of 1-2 devices.
Certainly the #1 factor in buying a second hub is that shipping time for a replacement isn't zero. and zero is all many people are willing to have their house down for. We'll have to see the impact of the Migration Tool and having a backup in the cloud and a Hub in the hand will yield.
If you do have a spare hub waiting for a failure, and the failure is as rare as they seem to be, other thoughts creep in.. like splitting the house so that a hub fail affects just half the house. I do exactly this.. I have a hub for my 'upstairs' devices and another for 'downstairs'. Each hub is as independent as I can make it. Motions often flip switches or dimmers, etc and I have ALL of that on a single hub, to the extent possible.
Time will tell if you alter your opinion of multiple hubs, but for those of us that have them, generally we've done so for a very good reason and aren't going to go back. If we have too many hubs, they seem easy to sell off to get down to the perfect quantity.
I have 6 hubs. Three of them are my primary set. I've mentioned "upstairs" and "downstairs" but the third is where all the Internet facing products reside. The few $$ spent on an additional hub to run Amazon Echo, Google Home, Weather, Dashboards and have a 'master for Modes or HSM' makes incredible sense to me, as well as being practical. We have no qualms about looking at Intel's CPU line and saying 8 cores are better than 4... yet when I do the same with Home Automation (3 hubs are better than 1) there's some flaw in the logic? Not for me. 
Maybe I was 'lucky' seeing many sales on Hubitat Hubs over the two years I've built my system, but it's likely I have less than $250 in Hubs which is roughly 6-10 ZWave devices. (I'm mostly a ZWave home.) The math is simple.
2 Likes
That's probably too broad of a statement considering all of the variables that go into a hub's performance (known and unknown). For example, like @Sebastien case, a problematic app segregated to another hub could make the rest of your home faster.
Unless you have a site to site VPN between your house and your parents', I don't think HubConnect would work. However, MakerAPI could probably be used... even simpler would be IFTTT.
False
HubConnect was built with this capability from the earliest days. Mostly because the Author goes Camping for months at a time and he has Internet in the Camper and ties it back to his primary Hub. There's a hundred plus devices in the camper... every door / window has a contact sensor so he can tell at a glance if the camper is road ready, as an example.
HubConnect offers an Internet option for connecting hubs:
The "internet" option is an outgrowth of the SmartThings integration.. The way into SmartThings is via their cloud. Well the same is true externally via Hubitat's cloud.
5 Likes
I stand corrected. 
2 Likes
Mostly false.
HubConnect processes events. It gets them from one hub and injects them to the connected hub. Hubitat Hubs are 100mb full Ethernet and compared to Zigbee at 250kb (yes, k vs m) the delay to transfer over the LAN is minuscule. If you use the EventSocket connection, there's zero additional processing on the sending hub.
If one hub (A) has two Z-Devices, a Contact and a Dimmer... and the desire is to have the dimmer come On when the Contact is open, and the Dimmer Off when the Contact is closed. Moving the 'calculation' (detect the event, determine that it has an effect, sending the action) to another hub (B) allows hub A to process Z-Radio queues. Z-Radios are Half Duplex. They must complete a transaction before beginning another. The result is that we want the Hubs to be idle... ready to process the Radio traffic, not off bringing in 2k worth of weather data. That desire to have idle hubs, ready to manage the slowest traffic we own, is a big portion of the benefit of multiple hubs. In a real sense, we're deciding to "dedicate a processor" to the Radio.
2 Likes
I do have a site-to-site vpn actually. lol. Everything is "local" to me. It makes it a lot easier to troubleshoot things at their house, even though they are just a few miles away.
Thank you for the info! I'm also checking out that RV page you linked, which just from the quick scroll is awesome.
There's a HubConnect specific Thread with over 3200 posts.
Read the first 3-4 then skip down to maybe the most recent 200. It will give you a flavor for the way it's being used. There will be a learning curve but there are drawings and videos to help.
And remember, I'm not suggesting you need more than one hub, but when YOU decide you do.. here's where the help resides
7 Likes
For sure! Definitely makes things easier from a support standpoint.
Smart bulbs are notoriously bad repeaters. It was more cost effective to add a dedicated hub for the lights than it was to replace all the bulbs.
4 Likes
Decided to take the plunge, ordered my Hubitat yesterday!
2 Likes
Same here! Bye Bye, ST.
2 Likes
I bought mine in June 2018, having had X-10 then VeraPlus. I was never happier both when I first purchased it and today.
- The hub has awesome support
- If you want to set it up and just have it do its thing you can
- If you want to be creative you can do that too.
I'll be purchasing a C-7 after the dust settles.
1 Like
For the sake of keeping the record straight, one of the key reasons that some have more than one hub, is because there are certain devices (e.g. Xiaomi sensors, Zigbee bulbs, etc.) that do not play well with other devices on a Hubitat Hub. In order to handle these unique and limited circumstances, many have decided to put these "problem" devices on their own hub, and connect everything together with the HubConnect package.
Many have also decided to have a development hub(s) and a production hub(s).
If you think about it, they've made the decision that having an "extra" hub is worth it.
1 Like
True Story :: Lost my internet for a day... all of my automations fired away no issues ... loving the platform
7 Likes
As a developer (for 30+ years!) and new owner, I greatly appreciate what Hubitat has accomplished! I have Alexa, and had SmartThings! HE is a breath of fresh air. As a former Java developer I have no trouble handling Groovy. I love the way the community is able to jump in and code their own drivers and apps. This highlights the great design of HE.
I am having a ball.
To the point of @SmartHomePrimer, can someone explain how adding another HE helps instead of say, using a zwave repeater? I am thinking of how I just added a couple aeotec zwave repeaters , one to a shed where I do have a hard wire. Sounds like a second hub would be better?
Anyway, great work!
We can offer things to think about, but ultimately you have to decide on "better". Now, I am very biased. I have 7 hubs. But only 4 of them are 'in production' the rest are Development.
I have a hub for "Upstairs" and a second for "downstairs" and a third that acts as the center of the collection. It also handles all the LAN and Internet devices: Weather, Alexa, Google Home, HomeKit (via Homebridge) and so on. My thinking was to treat "upstairs" like it was a 2nd home... if that one Hub faults, only that portion of the home is affected. Same with "downstairs". Those hubs are ONLY Z-Device hubs, nothing else. My thinking there is that the Z-devices are SLOW relative to the LAN. Anything I can do to keep them working on clearing any backlog of device commands, is a benefit. I want radio queues to be as close to empty as practical.
I've recently started adding in a 4th Hub to be yet another Z-radio hub. I'm calling it "Front" because there's a lot of devices in my home along a region at the front of my house and I'm thinking I can move a subset of 'upstairs' and 'downstairs' to (again) reduce the queue depth. It's a bit of an experiment because I have nothing that is needing this.. therefore it's a bit hard to measure 'success'.
Others have a 2nd hub just for backup.. to reduce the shipping time to zero, mostly. But why have it sit in a box?
A Hub in your shed would be useful if the count of devices in the shed AND their response time, was going to justify the Hub's cost.
3 Likes
