Right, I’ll continue to follow the thread !
As I already reported (in don’t remember which other thread), I already upgraded to 7.30 three Vera Edge (mine and my two daughters) without problems. Only several devices (Neo Coolcam sensors mainly) remained “waiting for configuration change”, simply “waking them up”, everything came again OK.
Regards
I upgraded my Edge with no problems. Took a little longer than usual, about 10 minutes with a couple of reboots. Feels a little faster but not as big a change as when upgrading to 7.29. Only issue I’ve found so far is that the City selector in the Unit Settings - Weather Settings section appears to be broken for UK - England. Selecting other regions in the UK works fine but select “England” and the City menu is blank. About 70 devices, real, virtual and Philips Hue.
You need to run @rafale77 scripts and you’ll get a more Stable and performant zwave network. Overall, there’s a lot under the hood.
1 Like
nope, I’m referring to those:
do not use the latest one (ReloadOnTimeJumps) 'cause it’s not intended to be really used on a production system (you’ll loose all time-based scenes).
Can someone simply explain to me if it is good to have nightly heal enabled (default setting) or set it to disabled or to be performed 1x week (as per Luup Requests - MiOS)
Thanks!
It depends a lot on your setup. There are “corner cases” where automating the heal could be helpful.
For example is if you have a small network (<20 nodes) with AC powered devices moving around your home regularly disrupting your mesh.
I see no good reason otherwise to have a nightly heal on a perfectly healthy and working network. The heal itself is a repair step which takes down portions of your network for various amount of time with the outcome of potentially breaking your network if you have FLIRs like locks and sirens run out of battery or with weak batteries. For general use, my opinion is that automating the heal is idiotic and dangerous. It is like going to a doctor for a heart surgery every day because you lost a hair a day. It is an operation which should only be run manually under very specific condition and controlled by the user.
Thanks, I have no devices moving from place to place, they are static. I have lots of battery operated devices and also lots of AC powered devices.
So it is better for me to disable the healing process using the URL address from Wiki.
If I am right, I still need to specify schedule of healing process in that URL address?
And if it is disabled, can I somehow run healing process manually? For exampel if I would move some devices from place to place? Thanks
I would disable it and yes you can still run it manually under the zwave menu. The heal is triggered when you click the “update neighbor nodes”.
The heal process is at the lower level is just the zwave chip of the vera sending an nnu (neighbor node update) command to all nodes from node 1 to 232 in that order and waiting for the nodes to respond. If a node does not respond (often due to conflicts caused be the formerly triggered wakeup nnu and wakeup polls or because of you sending a command from a scene), it will re-queue the command until it gets a response and that’s how for my system which has over 140 nodes, I ended up with a zwave network which was healing forever, starting a new nightly heal process before the previous one completed and eventually the whole system crashed. That’s why some members here are rebooting their unit every day and found it working better that way. Please refer to the thread I posted called “zwave network explained” I linked several times on this thread (and so did @therealdb a couple of posts above) for the detailed explanations.
2 Likes
They are lua scripts/lua codes which just add or change device variables or vera attributes. They are not the shell scripts I posted about extroot indeed.
Also good reminder from @therealdb on the time jump one. Unless you do not any time based scenes, do not run that one. People who use openLuup for example to their time bases scenes can run this.
In thread about the Beta release the new location for device icons was stated to be /www/cmh/skins/default/img/icons/ . But that location - at least on my VP - is read only. Is that still the planned location for device/plugin icons?
Since you are a developer you should have access to the developers corner section? There is a discussion there about the 7.30 icon path change. Can PM me if you can’t find it.
OK the actual location is /www/cmh/skins/default/icons/
And the suggested reference in JSON files is "…/…/…/icons/<your_icon_file>
Thanks, these are the ones I understood, BUT my question is: Re-reading @rafale77 post about them it seems that the effects they were intended for were already implemented in 7.30, so, no need for execute them.
So my question is, even having upgraded to 7.30 should we still execute them (except the last one (ReloadOnTimeJumps) to avoid loosing the time driven scenes? Or 7.30 already implemented what they are intended for?
regards.
I see, I didn’t realize this was causing confusion.
The lua codes only work with 7.30. They do not work with any other firmware.
They are not enabled by default so yes you need to execute them.
I think they only have an impact on 7.30 so the ability to change them is implemented in 7.30 but the default value is not what we want…
C
Understood, thanks for your such a fast answer…. I will do…
Javier
Thx, I’ve already tweaked those settings so I’ll see how it goes.
Thanks guy’s for the positive reviews, my finger hovered over the Upgrade button for a while, I closed my eyes and and hit Upgrade wondering why I do these things…
At the 20min mark it was still Updating.
At 25min. Power cycled Vera
Now I have a borked Edge…
Why do I do these things I hope CS is awake.