[RESOLVED] Unable to create new scene with Xiaomi button after update 4.5?

Hello everyone!

First of all, thank you for the Gladys update, it is now possible to control multilevel devices in scenes and that’s cool because I get a function that I use every day!

Thanks again for developing this feature!

So I started setting up my scenes to turn my lights on/off via the action of my Xiaomi buttons, but… Impossible to set them up…

Indeed, it seems that since the update, I am unable to create a new scene with my Xiaomi buttons as a trigger… (Ex: Click on Xiaomi button = turn light on to 100%)
However, the newly created scene works fine when I click on the « start » button once the scene is saved (the multilevel therefore works perfectly), but not when pressing the button.

So I thought the problem was with the button, so I tried to create a scene with a button already configured in Gladys (to control outlets and which therefore works perfectly…)
Well, it doesn’t work either… (Same, works when I click on the « start » button)

The scenes with the buttons as triggers that were already configured before the update work perfectly. And I can even add a secondary action within it and it works perfectly because the action added to the existing scene is launched when the button is pressed.

But when I create a new scene with a button as a trigger, nothing happens when I click on the button.
And this regardless of the action I want to launch from a Xiaomi button (opening shutters via http request, turning light on via the new multilevel feature (Zwave), or controlling outlets (Wifi))

Any idea where the problem might be coming from? Because I’ve been trying to find out where the problem is for an hour and I’m stuck…

Hello @Retlaw!

Cool! Glad to see you back with us with v4 :slight_smile:

I was able to reproduce the bug at my end, it is indeed a regression introduced by v4.5 following a bug fix (Fix bugs in scene, dashboard & settings (#1228) · GladysAssistant/Gladys@183742e · GitHub)

I just created a PR that fixes the bug:

https://github.com/GladysAssistant/Gladys/pull/1276

I’m trying to merge this quickly so that it goes out in a bugfix release!

Sorry for the inconvenience :slight_smile:

Update: Gladys Assistant v4.5.1 is currently being built with the fix :slight_smile:

https://github.com/GladysAssistant/Gladys/actions/runs/1204443890

It’s done, Gladys Assistant v4.5.1 is deployed and available!

Your instance will update automatically within the next 24 hours :slight_smile:

If you want to force an update yourself (this requires CLI attention, only do it if you know what you’re doing. Gladys will update itself anyway), you can run Watchtower in « update all containers now » mode with this command:

docker run \
  --rm \
  -v /var/run/docker.sock:/var/run/docker.sock \
  containrrr/watchtower \
  --cleanup --include-restarting --run-once

Hello,
Just for information, do we run this command from PuTTY?
Thanks for the project progress.

Are you joking? You don’t have to apologize for that! You’re adding the requested features to Gladys, what more could you ask for?

Given the work done, and the price (exorbitant!) it costs to use Gladys ( :joy:), it would be totally inappropriate to complain about anything!

The community is there for that, to report bugs and move the project forward! If I can contribute my little brick to the edifice by reporting them, I do so willingly!

And it would be even more out of place to complain when you see your reaction time! Less than 24 hours to resolve a bug and make everything work! Hat’s off!

Thank you for your responsiveness!

Edit: By the way, I am very pleasantly surprised to see that the V4 responds to the 1/4 turn, much less latency than when I was using the v3!

Aha, thanks that’s nice :smiley:

Thanks for your feedback anyway, and this kind of positive post, it’s not always the case in open-source and I much prefer when it goes like this :joy:

That’s the goal, there are many deep changes in the way the v4 works compared to the v3 that make it normally much smoother to use.

Don’t hesitate if you ever have other feedback on the v4 (small missing features, details that annoy you in the v4…).

And by the way, I take advantage of it now that you have switched to the v4, I don’t know if you have already tested Gladys Plus, but I recently added a completely free 14-day trial offer (no credit card to enter), if you ever want to test the encrypted remote access / the open REST API / the Owntracks API, and soon the Google Home alpha, I would be happy to have your feedback (it’s completely commitment-free) to be able to improve the product!

If you want to test it, it’s here: Gladys Plus | Gladys Assistant

Yes it’s to be executed in SSH on the Pi if you know what you’re doing.

I specify, I gave this command here exceptionally because @Retlaw is a former user of Gladys v3 and he knows what he’s doing, but Gladys updates automatically so no need :slight_smile: .

The philosophy of the v4 is really not to have to use the CLI so it’s not a process that I want to document, and it’s intentional :wink:

Oops, the answer is there Gladys 4 - Développement du service Arduino - #144 par JPL

I saw that the 14-day offer had been put in place yes.
I will test it with pleasure and give you feedback, but for now, I am going on vacation so it will be later :wink:

Exactly, it’s the perfect opportunity to test remotely

Sounds good, and if during your vacation you are available to follow the YouTube live stream, I will do a demo of Gladys Plus (and technical presentation) this Thursday at 9:30 AM :slight_smile: