The answer used to be a solid no every time the question was asked if yaml was going away. Now it is a “No…but” Sounds like back tracking to me
Oh look, a user who is screwed because of a UI configured integration. This would have been dead simple to fix with a yaml config.
Private github repos are still readable by github.com employees. You know, Microsoft. So yes, good that it’s not a public repo, but realize you’re still trusting outside your zone of control.
Other git based repos are avaliable. I quite like BitBucket myself
You can have a local git server also, some people use git on synology NAS for example.
That should be “a second user…”
I’ve asked twice now in this thread (three if you count this time) but I still haven’t had anyone who promotes the idea of “UI only integrations” explain how someone is supposed to “officially” fix a failed integration that was configured via the UI if the UI doesn’t load.
Come on, people, let’s “Make It Easy™”!
I have yet to have an integration that a configuration issue caused the UI not to load.
The user in question in that thread had a custom component who’s dependency failed to load, not sure why THAT caused their UI not to load but since it’s not an official component/integration who knows.
Maybe you haven’t but I did. And it took several hours of troubleshooting to get everything working again. If I was a “non-tech savvy” user I wouldn’t have been able to even come close to being able to recover from it.
And how common it is isn’t really even the point. It can happen.
But even if the cause is something else (custom component or not), if the error logs say that the error is caused by a certain integration then the logical first step of any troubleshooting procedure has to be removing that integration. Otherwise the error logs are completely superfluous.
And if you can’t remove that component (custom or not) then you are still in the same boat in the end.
There has to be some official way to recover from a failed integration/component configured via the UI that is just as “easy” as simply removing the configuration from your yaml files.
Not to mention that a number of custom components are moving over to the ui set up, so we’re still in the same boat. I’d like to think that this means an ‘official way to fix’ is on its way, but I suspect in the long run they’ll just remove the ability to add custom components.
Yet more:
tell me about it. I am stack from the morning and now I am looking for a way to reinstall HA and I don’t know how
I just had an epiphany…
We are all thinking that the “Make It Easy™” concept was to make it easy for users when all along it was to meant to make it easy for the dev’s.
Now it makes perfect sense!
I keed, I keed…or not…
The same I said 12 days ago.
Of course.
Always late on the uptake…
What is amazing is how Home complicated Home Assistant is. It has over 1500 integrations, lots of custom components, and many systems that it will run on. The more you use it the more you find out what you don’t know. I appreciate all the effort that the developers are doing. Making it easier for development is a good thing. It is a long way from being easy for non technical users.
I think it fits here.
That’s what I suspect…take away all the customisability (not UI configurable) is the road they are on.
Your “more” example is another user using a custom component. Those components aren’t supported by the HA devs or the core system so their behaviour can’t be guaranteed. All of your examples so far have been people doing unsupported things and then having stuff break.
Can you find an example where someone wasn’t using a custom component?
9 or so posts up:
There’s no indication that user wasn’t using a custom component nor does the user indicate which component they were using. It’s hard to say whether or not it was a supported integration.
That actually is the point, if it’s “impossible” with a supported integration the the discussion is moot.
Custom components are an essential adjunct to home assistant. Many users have one or more custom components. The whole HACS system encourages installation of custom components. HACS is developed by ludeeus who is a major contributor to HA core (59th largest number of commits to core github out of 1906 contributors).
thomasloven is the 9th highest contributor to home assistant frontend and the 9th highest to hacs/integration.
Custom components are risky, but are hardly a no no.
If this is an attempt to squeeze them out, it is a strange way of going about it.