I’ve been a long time HE user. I cannot say I like it much as it requires significant efforts to do simple tasks but it work for in intended purpose. A few months ago I think after one of the updates the reliability went down the hill. It hangs without communication with the devices and unless you are actively monitoring it you do not know. Once you discover it is down the only way to revive it is to cycle the power. Then it works for a month or so and does it again.
It is a huge dealbreaker for me. I am seriously looking to switch to some other hub and the only things stopping me is that I am reluctant to find and reregister dozens of zwave and zigbee devices.
I would suggest the team to be more careful changing the core code. Maybe some more unit/integration/functionality tests would not hurt
All the new features do not matter if the hub is not reliable. Just my 2c
Then why do you continue to use it rather than find an alternative? I recognised the username and had a quick look through your posts and most seem to be complaining that Hubitat isn't up to scratch.
5 months on you post to criticise the product again. I just don't get it
did you post your issue when you had it? the team and community are very engaged and find solutions/workarounds, if not fixes/code upgrades to address the issue
MY hub is quite reliable. I may not have all the devices programmed to the extend you have set up, but I have made the hub barf or had a misbehaving device give me grief. It was very useful to find out why MY PARTICULAR setup did what it did, to set things straight. Up until that point, yelling at the moon got me nowhere, despite the effort.
Just saying. . .
We could not agree more with your statement. All of the updates that we provide are intended to make the hub more reliable, while offering new features based on the feedback we received from users.
If an update has negative effects on someone's hub, we also offer ways to restore the previously running version that may appear to be performing better on an individual hub. Plus, if a hub is running well, and you are not interested in the new features, you don't have to update. We never force our users to update their hubs if they don't want to.
Based on hundreds of incidents that we have reviewed, the majority of cases are related to changes that occurred within the local environment and not so much related to changes to the core code. Most common are changes within one's network structure or misbehaving devices. In very rare cases, the hub's hardware may be the problem.
Without additional details specific to your hub (including but not limited to your hub model, overall set-up, mix of devices, etc.) it is hard to know which category you may fall into.
It is very challenging for me to reproduce it. All I know that the same setup that ran for a few years all of sudden started failing. Is there some kind of dump solution that takes the hub state into a file?
@vadim_c
May I suggest a "temporary" band aid to get you going again?
There have been many reports in the past, of systems running out of useable memory, and requiring a reboot to get them going again.
(I'm not sure if you are one of these types of systems).
There are a a few approaches that have been taken by various people to get around this issue.
- A brute force approach, is simply to write a Rule machine rule to reboot your system every week, at a time when it will not impact anything. (There are many examples of this, please search the forums.)
- A more nuanced approach, is to check the amount of free memory that you have, and reboot when it falls below a certain threshold. Checking memory is done via the Hub Information device driver, and rebooting is also done via that same device driver. Please search the forum for examples.
Another very common issue has been the occasional power drop - which may cause Hubitat database corruption. This is mitigated by a "soft restore". Again, search the forum for exact details on when and how to do this.
Another common issue that I have noticed, is the occasional device which just goes crazy - i.e. it reports an abnormal amount of times, and it just "drowns" everything else out. Sometimes, this may be spotted by a close examination of the logs. Yes, it's true - devices can just go "bad" (sometimes even randomly). These situations are difficult to diagnose, but, they do happen.
Many people here are experts and are more than willing to help you with your issues, if you give them enough information on the cause of your issues.
While you may lose features and lose some functionality (newer device drivers, etc), you could roll back to a previous release to get stability until such time that you feel you can upgrade again. Nobody forces you to update Hubitat hubs, if something is working fine, you can just stay on that release if you feel it works for you.
There has been extensive testing for memory issues in the past, and now presently in Beta test group. There have been quite a few Beta versions in 2.3.7 trying to nail down memory.
You also could contribute by being proactive and trying to help staff and the community by testing if there is some driver, app, or something else entirely that is causing trouble on your hub. Many have joined Beta testing to test potential fixes. Others who want less involvement just contact support, and have support look and see if there are internal logs/errors that indicate issues. Even working with community members here could help if you post requesting help.
Any clue to help staff find and fix issues helps everyone.
I have had similar issues over time. After a month or two the response will slow down and if I don’t catch it in time the hub will lock up. I did notice some delay today and just checked and my C-7 was down to 107MB. Not critical but I went ahead and rebooted all three hubs.
I have thought about doing a db refresh but haven’t looked into it. I’m in the process of a system rebuild and migration. Not because of HE but rather my system has gotten out of hand due to my own tinkering and expansion. Once I move everything off the C-7 it will get a full factory reset.
I have no complaints about the system. Just noting that I have seen some memory related issues. No sense in chasing them down since I’m starting fresh and will be wiping the hubs.
I agree with most of the posts here. The community (users and HE staff) has been very supportive.
Just me, but I probably would do one prior to migration just to be sure I didn't have any issues. It only takes a couple minutes to do this procedure.
I’m not doing a migration. I’m moving devices one at a time and rebuilding everything.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.
