I would suggest to customize month start for the new energy page. Not all energy providers starts month costs calculation on 1st day of the month. in my case my consumption month starts on 16th of each month. the way system works now, i cant evaluate my expenses to current month as system calculates from 1st to 30th, which does not correspond to cost calculations for those people. would be nice to implement this config.
Not sure why you are replying to me.
Just my opinion. Working on a raspberry PI, homeassistant==2021.8.8, everything was working great until I attempted the following update, sudo su -s /bin/bash homeassistant , source /srv/homeassistant/bin/activate , pip3 install --upgrade homeassistant. After attempting my update my Raspberry PI would not boot into Home Assistant. I reverted back to 2021.8.8 with the following commands, sudo su -s /bin/bash homeassistant, source /srv/homeassistant/bin/activate , pip3 install homeassistant==2021.8.8. and everything is back to working, I know I should follow the old view, “If it is not broke don’t fix it”. I know my system is still working on the [Raspberry Pi all-in-one installer and I am sure there will be remarks that I should upgrade to the latest’s version of HA. Sorry, I have tried that but my many modifications to my HA never seem to transfer to the latest version of HA!! I will keep trying.
Then it is even more surprising, and concerning, that the 2021.9.0 was released with this bug still there.
Not if it wasn’t reported
it obviously was
Then it wasn’t the same boot loop problem and your comment is a moot point…
Yeah so there were two pi3 errors with exactly the same symptoms in one beta cycle?
I can accept that is possible, but frankly you seem to be guessing.
I’m going off what dave said, it’s simple logic. Found a bug, applied a fix, the fix appeared to work. Again, according to Dave. But your what aboutism comments with all the doom and gloom are not welcome. It’s pointless fear mongering. “Oh man, it’s so so troubling that a bug got through”. Give me a break.
Hi
I have updated to 2021.9.1 and lost Smarthings integration. I have tried to set this back up but I can not pick my location in the setup. The drop down instead puts in the token so the setup completes but with no location.
Is this just happening to me?
Thanks
Andy
Added an image. The My Status should not be there in the UI and this prevents the Location being picked.
There have been many inquiries made in this thread about long term statistics and no answers yet.
Does no one who reads these threads have any idea how the new long term statistics work?
No you give me a break. You know I am a staunch advocate of HA and the development process.
I am just puzzled that the users are being told that we need more beta testers, yet when a beta tester appears to have caught this bug, it still seems to slip through. So why doesn’t someone tell us what happened. Was it the same bug?
I’m in Melbourne. Who are you getting this data from? Our water meters are still read manually.
The release bug was a RPi3 one in the final release - different to the B2 bug althougt both were boot loops
Thanks for confirming Dave.
@petro tagging you in light of all the chatter around bugs, betas, alphas and testing, which seems like a hot topic on this thread you’ve been part of (happy to continue this conversation under a new topic too), and since you’re a great contributor to this community with a deep understanding of it all: Do you have any advice for setting up a staging environment if you don’t want to use your main (production) instance for this purpose? I would think one would need a more vanilla setup, because you don’t necessarily want to have devices integrated with multiple instances, and have the same automations and notifications firing. It’s either that, or have a few spare devices (an old phone, a light of a particular brand, etc.) that then forms part of a staging environment. I would think the latter would be preferred, because most of the time the issue that creep up are related to integrations.
To all the complainers:
- Read the release notes. Always. All of it. You may not understand every bit of technical detail, but you surely can follow “I’m using integration x and the release notes mention a change to y in relation to x, so maybe I should be careful, test or hold on”.
- When there are major new features introduced, hang on before upgrading. Early adoption comes with risk. No amount of testing will catch all the cases. Sometimes big things will go wrong — even the biggest of corporates make huge stuff-ups (go look at AWS, Cloudflare and Google over the past year) with all their robust processes, tons of people and resources.
- Nobody deliberately is trying to upset you with bugs. It’s not personal. Consider the thick skins the core devs need to have when battered on the forums after a ton of hard work. Don’t be a passion killer. I can totally understand that issues sometimes get locked, especially once it’s fully understood. Another “me too” message certainly isn’t helpful at that point.
- No new feature is fully developed right at the start. Asking “why wasn’t x done” or “how could you (not)” questions are useless and comes across as entitled. Show a bit of understanding and consideration. These features take time.
- Run supported hardware. Use approved integrations (with active communities). And yes, of course those could break too — not interested in moot points; it’s about decreasing the likelihood of something breaking. If you hack something, accept the risk that comes with it.
What do you mean by conflicts in the network?
Depending on how you run deCONZ you can have multiple instances of HA talking to the same deCONZ instance. Thats how I work with deCONZ at home. In my case I run the standalone docker.
I know an ex ‘meter reader’ in Melb. As you can imagine some meters are harder to read than others, think business sites, difficult gardens, fences, etc. I believe ‘smarter meters’ are available to be installed for some customers. Mind you , never seen one or know how they work.