Remote.notify

[tinu]

Hallo zusammen

ich habe in einer Installation eine Alarmanlage in LCN direkt programmiert. Es sind diverse Reedkontakte plus ein BMI, welche den Alarm (Modul 204, Relais 1) auslösen können. Mit C.Remote.notify sende ich die Meldung "Alarm Einbruch" an zwei iPhones plus ein iPad. Alles funktioniert einwandfrei bei normalen Alarmauslösungen. Leider kriege ich ab und zu Fehlalarme (nur Meldungen auf den iPhones und dem iPad, teilweise auf allen Geräten, teilweise nur auf einzelnen). Die Sirene, welche vom Relais 1 des Moduls 204 ausgelöst wird, bleibt bei diesen Fehlalarmen stumm.

In der angefügten Datei ist der Event für die Remote.notify Meldung sowie ein Auszug des LCN-Busmonitors zur Zeit eines Fehlalarms zu sehen. In Zeile 53 wird das Relais 4 des Moduls 204 auf 0 gesetzt, nicht aber das Relais 1.
Wo könnte der Grund für diese Fehlalarme liegen?

Herzlichen Dank für eure Unterstützung.
Martin

[DOMIQ-Support]

@Martin. I have a question regarding notifications - are they always delivered to all devices when the alarm goes off? Maybe some devices don't get it on time and in result they are delivered with delay? As a debugging tool use this function:

Code: Alles auswählen

time = use 'MEM.trigger'
function triggerTime()
   time:set(date("%H:%M / %d-%m-%Y"))
end   

This function saves date and time to the MEM.trigger variable. Use it in your event for notifications (add as another command). In the Name field enter: C.LOGIC, in the Value type: triggerTime(). Now each time relay no 1 in module 204 changes its state it will saved in the MEM.trigger. Then you can compare it to the bus monitor logs and check whether this relay really changes state.

[tinu]

Dear Piotr

thank you very much for your help. I implemented your function as instructed and got another false alarm at 10:46. It arrived all three coupled devices (2 iPhones, 1 iPad) imminently at the same time. The LCN-Log shows at this time indeed relay 1 of module 204 at state 1. Before the time the alarm went off the same relay always shows state 0. So this must be for some reason a wrong reading of the status of the relay concerned because the sirene which is connected to this relay never went off. The attached file again shows the Log and the value of MEM.trigger.

Is there any known reason for this behaviour and how can it be supressed?

Kind regards
Martin

[DOMIQ-Support]

Well, if LCN logs also shows that this relay was on, then it was actually on or LCN sent wrong status. Base will never react on event that didn't really occur. However, if siren didn't go off, then I think it's caused by some wrong status message from LCN module.

[tinu]

during further troubleshooting I noticed that the wrong status of relay 204 was sent by LCN always just after the Domiq commands to relays 200, 201 and 206 to drive the shutters. I noticed this by receiving the notification to my iPhone right at the time the shutters drove down. The LCN-Log confirms this correlation (see attach of Nachricht# 53237). After deactivation of the event to control the shutters no more false alarms were produced.
Is there any solution to suppress these wrong status readings?

The attached file shows my event to control the shutters.

Kind regards
Martin

[DOMIQ-Support]

So you mean, you deactivated the event in Base and the problem dissapeared?

[tinu]

Exactly, no more false alarms since I deactivated the events in the Base which control the shutters.

[tinu]

Since deactivation of the events for the shutters not only no more false alarms went through to the iPhones but also the real alarms don't arrive the coupled devices! The events for C.REMOTE.notify was always correctly executed which is confirmed with the MEM.trigger variable. The alarms haven't come through now for more than 48 hours since the first alarm went off for test reasons.

[DOMIQ-Support]

Is the triggerTime function called properly? If it does, then I don't see any reason why C.REMOTE.notify isn't executed.

[tinu]

Yes, I think so. I didn't change the event for the alarm notification. I just deactivated the events for controlling the shutters and since then no more alarms are coming through to any of the three devices. The alarm event was always executed when testing. Again this is confirmed by the correct time of the MEM.trigger. See attach. My last test was yesterday evening at 20:05.