I've had a C8 since release. I haven't touched it since. Haven't added anything or removed anything nor changed anything. All I do is the Firmware updates that it.
Now It continually locks up about Once a Week I have to unplug the power and reboot it that way. It's really causing an issue in my house as in I take cold showers because its Locked up. Hub shows Green.
It's so bad. I ordered a plug to plug the hub into so I can tell Alexa to reboot it since Habitat is dead.
I realize its out of warranty, but I need some sort of solution. I've done reboots and soft resets and it still locks up.
I've had 3 or 4 C8 outages in the last month or so. Generally latest hub firmware or previous version. Total network outage (no access to diagnostic endpoint, no ping). Connected via Ethernet. Simple zwave rules ( eg button press -> light on) fail, Nothing recorded in the log. So it's not just a network issue but appears to be a full on crash. Power cycle is only recovery that worked for me.
Only significant change has to enable Matter support and the addition of one matter bulb - a Tapo L535.
Hmmm... I just realized thetr may be some synchronicity here. I guess I know my next debugging step.
Aside from power cycles, have you tried shutting down the hub, pulling power, waiting 30 seconds, and then restoring power.
That is the only way to clear the radios on the hub. If you're doing a quick power cycle, i.e., power pulled for just a few seconds, that probably is not clearing the radios.
I'm going to be honest and give you my opinion. I know people will disagree, But I purchased the C8 at release and VERY shortly after that the C8 Pro was released I remember a lot of people were not happy. I honestly think they figured out that the C8 was not able to handle what they wanted to do with the Hub and adding Matter. The only thing they changed was the Processor and Ram (From what I read). So this lockup issue I'm having seems to be getting worst after each update. I feel its running our of space? or the processor can't handle larger setups and I notice people saying to turn OFF Matter this is what caused me to look more into this. I don't think the C8 can handle much more.
They will never replace these C8 hubs that are causing issues and I'm NOT buying a C8 Pro. I was really not happy when they released it after I just bought the newly released C8. I have about 80 Devices and many routines and schedules.
I setup a plug to Alexa and having my C8 hub reboot every morning at 3am and I stopped doing updates as I honestly think this causing the lockups to get worst.
So Hubitat can send me a C8 Pro or I'll stick with daily reboots until I work on switching to Home Assistant. I really feel I should of been offered to upgrade my C8 to the C8 Pro for a small price. Like I said I feel they knew there was an issue with the C8 on large networks and adding Matter killed it. I also am going to cancel my yearly subscription.
That is likely going to make things worse. If you want to reboot daily you should use the Hubinfo driver. Monitor Memory and CPU and if mem gets excessively low, or cpu gets excessively high reboot it cleanly from the HubInfo driver.
Pulling the power out from under it every night will just cause it to have issues.
The thing is no matter how you spin this, any of these hubs have limited resources. If you had a C5, C7, C8 or the C8pro the resources are not that hard to overrun. They are tiny machines, and we ask them to do allot. The C8 Pro got a new SOC that only gave it another 20% or so of performance. The memory is probably the bigger deal, but that is more about memory optimization. That unfortunately can be environment dependent.
It is very possible you have a missbehaving automation or rule that is creating to much load, or is combining with another task that is causing conditions were tasks that can't complete and build up to eventually cuase a lockup. Have you tried turning off automation and rules to see if you help with it.
Though your use case may be an example of a problem, I had over 150 devices configured on my C8 with a fair number of automations and such running on it. It would run for weeks generally fine. Was it low on memory resources yea, and my rule with Hubinfo that monitored memory would occasionally trip to reboot the hub. I did move to my C8 Pro as a way to mitigate it.
It is also very possible if you bought a Raspberry pi with 1G of ram and loaded HA on it you would have a very similar result. Though the good thing is now if you did go that route you can get a Pi4 or 5 with much more ram so it isn't really a fair comparison.
The key take away though is that all of our environments are unique and the rules we want to implement are unique.
That isn't true. My C8 runs Matter just fine. The reason for the C8 Pro is likely the same reason we went from C4 to C5 to C7 to C8. There were changes in availability of chips and chipsets over time. Hubitat probably contracts to purchase X number of hubs with a particular set of chipsets, and it doesn't make sense to make a run of older chipsets or they simply are no longer available. It also makes sense for marketing purposes to create a higher spec hub.
These hubs were released a year apart, with different specs, which is no different than most any consumer device.
I bet the next hub (C9? C10?) comes standard with the C8 Pro memory specs (or better). It is just what raw parts you can buy in 2024 and beyond.
That isn't true. I have seen support replace numerous hubs over the past 5 years that were shown to have hardware issues. On the other hand, if there is a spammy device, bad community driver, or bad rule that is taking down a hub, they won't replace the hub.
They aren't going to just send you a new hub just because.
You can feel all you want, but this isn't true. You need to work with the community and or support staff to figure out what is going on with your situation. It might be hardware, it might be something totally unrelated. I have seen things on this community like people putting their hub on top of their Wifi router and wondering why it didn't work. There are numerous devices that are known to take down a hub. Simply upgrading to a C8 Pro is likely to have the same outcome, a hub that locks up. Maybe it will go 10 days instead of a week, but that isn't solving the issue.
@eastpascoguy I moved all the side convo out to another post.
I see this got posted in the Lounge section. Are you looking for help or did you just want to vent? Thread could be moved over to "Get Help" if that's what you want, so it is more clear.
Some people come on here just to discuss but dont actually want any help in figuring it out. Which is OK I guess, but people will instinctively try to help anyway.
Ok lets start over here with some real troubleshooting. I read through the thread and things did not really go anywhere thus far.
The first thing to establish is if the hub is actually locking up or is it getting booted off the network and not reconnecting.
For sure try the diagnostic tool next time the hub is offline.
It would be good if you can try and ping the hub when it goes down.
And then once you get it online again check the logs, and see if there is any log history while it was offline or is there a huge gap in the logs?
I think I would like the following info (hand picked from my read first post):
Ok. I'll try and ping it next time. I know it will not load via website. I took it off daily reboots as someone mentioned it would cause issues. I do honestly feel its getting worst after each update. When i see other people having the same issue it just seems like something is going on. If it was only me I'd say i have something going wrong on my end.
Just an FYI. After an update the issue went away. I think they knew there was an issue. Like I was saying. I had no issues of lockups until after an update. I think it was 2 updates after I posted this the issue was fixed. I bet they quietly patch the issue. I don't see anything in patch notes. But trouble shooting was easy... Before update no issues... after update lockups.
So you're assumption is that the most likely explanation is that HE staff did not want to get credit for working on and solving a problem for one or more of their customers. Instead, they chose to hide/sneak it into a release and never talk about it. That makes sense?
If so, I assume HE will stop posting any info in platform releases about bug fixes, since that would make their customers as happy as can be, because they'd be confused and unable to determine if their issues were getting any attention at all, or had been fixed or not. Brilliant approach?
Are any of below is just too wild to consider?
Something you've changed in the interim that seemed unrelated, actually resolved it
Some change in the platform updates resolved it, unbeknownst to HE
Some combination of both of above
HE did actually put in something to fix it but it got lost during the release process and didn't get into release notes
I've seen all of above over the years working on SW/FW. Seriously, no need to turn this into a conspiracy...sometimes you just get lucky. Why not enjoy it?
