I feel your pain. I sincerely hope your new C5 works better for you than your C4.
1 Like
Watching the thread as I would still love to understand what happened to my hub and if it's possible to bring it back to life but I'm done spending hours of time on troubleshooting to be told it's my network.
I had network issues with mine. I ended up buying a TP-link 5 port unmanaged switch for $15 on Amazon and hooked up my old SmartThings hub, 2 HE hubs, and a PC running HomeBridge to it and everything's running well so far. It's only been about a week, but I was able to pause my nightly RM reboot rule that had been running for the past couple months.
@Ken_Fraleigh I have several switches, but I no longer have any old routers (only the one I am using). If you are using a switch, how are you assigning IP addresses without DHCP from the router? Did you set up a separate one?
It looks like Hubitat does not allow static IP addresses.
Also, would you mind sharing your RM setup for schedule nightly reboots?
With a complex system comes complex problems. And I have no doubts that there is a legitimate problem here. There have been enough reports of hub slowdowns and outright crashes (port 8081 not responding) by a wide sample of users.
There could be a dozen things that cause these symptoms but that doesn’t de-legitimize the complaint.
It’s unfortunate that support is not directly acknowledging the issue because it has reached the point were credibility is starting to be lost.
You can add me to the list of those who have been affected.
5 Likes
We believe there is indeed a resource leak in some systems. And, there have been other issues that cause slowdowns as well, including errant apps/drivers and a networking problem with some routers. We are well aware of the issues, however, it is not so easy to diagnose.
We are taking steps to find and fix the resource leak, and I can say we don't yet have a solution for it. When this happens, as some have found, rebooting the hub clears up the problem, at least for some amount of time. The dang thing is very sneaky and hard to pin down.
15 Likes
Prepare for this thread to explode.
3 Likes
Thanks for the info, I'm available to test new UI, to re do my entire home if it will help HE, I have 2 hubs in my house, 3 friends with HE that we can use for testing. Good luck catching the bug.
1 Like
There are multiple problems with the same eventual symptom: slow down.
One is the network interface issue that can be mitigated by 'autonegotiation' -- this is where the UI is pretty much always slow. Nothing else is affected. Automations run normally. Ideally the platform could/should identify high error rates on the Ethernet port -> another red ball warning, perhaps.
One is where the DB is corrupt. The corruption seems to apparently extend the DB response time and again, the platform could help by monitoring access times and --> another red ball warning, perhaps.
It's been several platform releases since these symptoms were identified. I think it's time the platform participated in diffusing the common eventual symptom. While it might be wonderful to never have a DB corruption, while Hubitat users wait for that eventual day.. couldn't a warning help?
In other words, people with those symptoms come here and get advice to get the autonegotiate patch or to do a soft reset, how harmful would it be to have the platform give that advice.?
4 Likes
I exhausted all of my troubleshooting options and got frustrated to the point where I switched platforms, so I can get where the OP is coming from. This stuff is supposed to be Home Automation, constantly babysitting the hub and worrying about keeping things running smoothly kind of defeats the whole purpose.
Good to see some official acknowledgement of this resource leak problem from Hubitat, and I would hope that more resources are dedicated to fixing this issue before adding more features. I know at least in my case, it wasn't the LAN autonegotiation, or custom apps - I had a mostly stock and clean system.
I loved Hubitat when it worked, but rebooting every night and slow response times is not something I want deal with at all. Hope to return soon when everything is running smoothly.
4 Likes
I set the static IP address on the router. It’s based on the device MAC address so it still assigns it no matter where the device is located on the network.
1 Like
@bravenel thank you for jumping in with your perspective on this. I'm glad to know that you are aware and the team is working on a solution...
I can say is that there are several of us here in the community who have strong technical backgrounds. Speaking for myself, and I am sure a few others as well, we are ready and willing to pitch in whenever needed.
I completely agree. I would offer the interim suggestion to create an official scheduled reboot feature. I have no problem with a nightly reboot.. I may have 5 hubs but we're not exactly talking a high availablility VXRails cluster. The 5 minutes a reboot takes would be far less painful than waking up/coming home to a dead hub.
7 Likes
I used the C4 hub (from last year to this summer) which would occasionally require a power cycle (stop responding). The location I was using this hub at has been under renovation since and will probably finish later this month. Should I continue to use the same C4 Hub, it is probably a few software versions behind as it has been sitting unplugged for a few month. Or should I replace it with a C5? Or preform a soft/factory reset?
I acquired a C5 Hub in October for another location (400+ miles). It has been stable since - been very happy with it.
I simply decided some months back to reboot my hub each night as per the example above. It is extremely easy to achieve. No addition functionality is required in my opinion. I have a small start-up RM/rule that reruns some initialisation of Chromecast/wifi devices etc to ensure they are reconnected. It works fine. While it will be great to get a proper solution for the obvious and ongoing resource leak issue, I always smile at the anxst in the forums about hub slow downs etc. The temporary fix is simple. And given the open nature of the platform, even if the underlying issue here is fixed, some dodgy app or driver can cause it again anyway. So unless you are an absolute purist running a vanilla hub, what's the big deal? Just reboot regularly.
1 Like
I posted above that I'm willing to help, I have 2 C4 in my house but they work fine, no daily reboots or whatsoever, but I have friends with C5 that those been slow sometimes. I suggest just stay with your C4, there is no reason to replace it.
2 Likes
Thank you! This worked perfectly!
Sorry, I misunderstood. I thought you meant that you replaced your router with the TP-link switch and thus no longer having a router for DHCP. I am using reserved IP addresses on my router to assign a specific IP to my HE based on MAC address as well.
1 Like
Agree. The only hub I have automated a reboot is my C5. My 2 C4s are rock solid. This said my C5 is my “coordinator” hub that has many custom apps and all LAN type integrations such as thermostats and AVRs. Only exception is HubConnect which is running in all 3 of course. Radios are off on my C5 too.
For the first time since owning HE I have held off on upgrades given what I have been reading about the latest versions. Running 2.1.5.124. This is the beauty of HE having the ability to choose when to upgrade.
@bravenel My development hub is a C3 and I am happy to load beta firmware on it and restore my coordinator backup to it to help validate any fixes. Please let me know how I can help.
it’s the auto-negotiating that isn’t working properly between Hubitat and some switches and routers. I’m not having any problems with the TP-Link TL-SG105E. It’s $24 on Amazon and I have 2 Hubitats, 1 SmartThings hub, 1 Hue Bridge, and Homebridge connected to it.
It's definitely some, but not all C5 hubs. Or it is much slower on come C5 hubs than others.
My C5 only gets reboot on firmware updates, and never any other time. Runs fine 24/7.
Both of my C4 hubs, on the other hand need reboot every few weeks or they get slower, and slower. But they have more apps and devices installed than my C5 hub, too, so could be an app/driver thing and not a platform problem. 
1 Like
6 days ago I moved all of my end devices along with 4 Samsung plugs, 2 Sylvania plugs, and 2 GE in-wall dimmers to a new HE hub. So far I haven't seen any slow downs on either hub. All of the lights are on my original C-5 hub, along with 6 Peanuts and 2 Sylvania plugs. I kind of wonder if it is a zigbee or rule maintenance/cleanup issue. I noticed quite a few devices, that were removed from all rules, still appearing to be associated with rules as I was deleting them from the original hub.
EDIT: I also had an issue with ML yesterday where I couldn't edit or create a new ML instance. Even after 'deleting' ML from the hub and rebooting, the logs were showing ML errors. I finally restored from an earlier backup and found a scene null value and fixed it. Then ML was okay again.
