All-in-One Scene: How to do it Better

[Rover 47]

18 minutes ago, cag014 said:

To all,

I'm trying to speed up command execution and to keep CPU low as much as possible.

Please try attached version and report if it's faster and CPU load still low.

 

 

Please login or register to see this attachment.

V6.1 and 6.2 are thinking that my D/W sensors (as SlaveSensor) are not breacheable.

Please login or register to see this spoiler.

Also  deleting the line with the rejected D/W sensor do not help, because then another D/W sensor is rejected...

[cag014 731]

35 minutes ago, Rover said:

V6.1 and 6.2 are thinking that my D/W sensors (as SlaveSensor) are not breacheable.

Please login or register to see this spoiler.

Is this sensor (168)  included in any alarm zone?

[Rover 47]

13 minutes ago, cag014 said:

Is this sensor (168)  included in any alarm zone?

Not yet.

[cag014 731]

3 minutes ago, Rover said:

Not yet.

So if it doesn't included in alarm, how do you expect to use "alarm" property... it's will be never updated or activated. The value will be "nil" and could cause unexpected failures (depends where used)

Edited October 27, 2020 by cag014

[Rover 47]

Just now, cag014 said:

So if it doesn't included in alarm, how do you expect to use "alarm" property... it's will be never updated or activated

At the moment I am developing/testing the AOQ alarm lines with two sensors. The rest of the sensors is getting activated in alarm zones when all alarm features are OK.

Meanwhile it would be nice if I can develop without a crashing AOQ initiation because of sensors not yet part of an alarm zone. A notification of a device alarm feature issue is enough.

[cag014 731]

I do understand that, but in any case it will not work since "alarm" state is "nil". All your lines with that could not be verified.

From other hand at normal operation I want to stop AOQ execution when using "alarm" property while this is not correct..

Any idea for workaround?

 

[Rover 47]

2 minutes ago, cag014 said:

I do understand that, but in any case it will not work since "alarm" state is "nil". All your lines with that could not be verified.

From other hand at normal operation I want to stop AOQ execution when using "alarm" property while this is not correct..

Any idea for workaround?

 

Yes, I could create a "Unused" alarm zone.

[cag014 731]

1 minute ago, Rover said:

Yes, I could create a "Unused" alarm zone.

Don't follow you... if in any case you add the sensor to "Unused" zone, why not to add it to "real" zone? In this case you will be able to verify that all commands are executed correctly.

[Rover 47]

Just now, cag014 said:

Don't follow you... if in any case you add the sensor to "Unused" zone, why not to add it to "real" zone? In this case you will be able to verify that all commands are executed correctly.

Because I will firstly test all alarm code, without complaints of my wife about alarms when opening doors or windows ?‍♂️

[cag014 731]

1 minute ago, Rover said:

Because I will firstly test all alarm code, without complaints of my wife about alarms when opening doors or windows ?‍♂️

That's the best reason to do it... never mess with your wife.?

[cag014 731]

@Rover

After receiving email from Fibaro that AOQ is turned OFF, could you please generate system report? Let's see if there is anything there that could point what the problem is.

Go to General

Please login or register to see this spoiler.

 

[cag014 731]

@Rover

May I ask you "stupid" question -

Could be that you have turned on notification when AOQ is turned OFF? The toggle button on hcWebEye turns off/on the AOQ to update the data table.

Please login or register to see this spoiler.

 

Just for testing, I've turned on the notification and have received same email fro Fibaro!

Please remove the notification

[Rover 47]

7 minutes ago, cag014 said:

@Rover

May I ask you "stupid" question -

Could be that you have turned on notification when AOQ is turned OFF? The toggle button on hcWebEye turns off/on the AOQ to update the data table.

Please login or register to see this spoiler.

No so stupid at all ?

The notification as you showed it was active. I think this is the default setting when generating the QA.

[gggizmo 6]

2 hours ago, cag014 said:

To all,

I'm trying to speed up command execution and to keep CPU low as much as possible.

Please try attached version and report if it's faster and CPU load still low.

 

 

Please login or register to see this attachment.

cpu load increased by 3-5%, but with more spikes than 6.1. I'll report you back about response time asap

[Rover 47]

Can I use any of the slave devices already there for a Fibaro smokesensor?

[Rover 47]

3 hours ago, cag014 said:

To all,

I'm trying to speed up command execution and to keep CPU low as much as possible.

Please try attached version and report if it's faster and CPU load still low.

 

 

Please login or register to see this attachment.

With v6.2 the reaction time beween pushing a button on Scenemaster and reaction on dimmer (on or off):

On: 2 minutes, 51 seconds

Off: 3 minutes, 21 seconds

So it do not look like as faster, but the CPU load stays about 20%.

I have to admit that meanwile I have optimized the jM: all long trueActs have been converted to lines triggered by device or GV with a minimum of conditions, 134 action lines now.

 

[cag014 731]

6 minutes ago, Rover said:

With v6.2 the reaction time beween pushing a button on Scenemaster and reaction on dimmer (on or off):

On: 2 minutes, 51 seconds

Off: 3 minutes, 21 seconds

So it do not look like as faster, but the CPU load stays about 20%.

I have to admit that meanwile I have optimized the jM: all long trueActs have been converted to lines triggered by device or GV with a minimum of conditions, 134 action lines now.

 

What!? over two minutes?

No way... is your HC2 reacts properly? Have found in some cases it's overload God knows with what...

Something very wrong here...

Do you have any device connected directly to HC3? Please try to use the SceneMaster to activate this device, interesting how long it will take

 

Anyone else has same reaction time?

Edited October 27, 2020 by cag014

[cag014 731]

@Rover

Could you post ScanTime and ScanMax values?

Please login or register to see this spoiler.

 

[Rover 47]

12 minutes ago, cag014 said:

What!? over two minutes?

No way... is your HC2 reacts properly? Have found in some cases it's overload God knows with what...

CPU load HC2 under 20%.

1 minute ago, cag014 said:

@Rover

Could you post ScanTime and ScanMax values?

Please login or register to see this spoiler.

I still cannot find these values. Where is this yellow rectangle in webEye?

[cag014 731]

2 minutes ago, Rover said:

CPU load HC2 under 20%.

I still cannot find these values. Where is this yellow rectangle in webEye?

Hover the mouse over

Please login or register to see this spoiler.