Thermostats - Desired temperature incorrect about homebridge-fhem HOT 17 CLOSED

are the values wrong directly after homebridge start?

please check if the homekit values are updated correctly if you use setreading to change the reading value:
''''
setreading MAX_03975d desiredTemperature 10
''''

please compare to the console messages and also to eve.

from homebridge-fhem.

Yes,

setreading MAX_03975d desiredTemperature 10

works as expected. That's strange!

Console output for setreading:

2016-11-01 10:40:08 caching: MAX_03975d-desiredTemperature: 10
[2016-11-01 10:40:08] [FHEM]     caching: TargetTemperature: 10 (as number; from '10')

But there is no console output for updated desiredTemperature by Max!
I guess homebridge does not realize that the reading has changed.

EDIT:
There is an update for the Max! fhem plugin. I will try if fixes the issue.

from homebridge-fhem.

Nope, the plugin update doesn't fix it...
(still no output on the console)

Changing the desired temperature by using the dropdown menu in fhem has the same issue. Only setreading works.

I guess it really is some sort of event catching issue...

from homebridge-fhem.

if setreading works and the other not then the homebridge side is ok and there is some event issue on the fhem side.

open two browser windows and check if one gets correctly updated if you use the dropdown in the other. also check if the event monitor shows the event and if the device detail view is updated (red timestamp).

if not: please report it as a bug in the max module.

from homebridge-fhem.

Okay, I will close this issue as soon as it is resolved.

if not: please report it as a bug in the max module.

See: https://forum.fhem.de/index.php?topic=60000.msg513461#msg513461

from homebridge-fhem.

I've updated the plugin to the newest version.
After further investigation I found out that change events are thrown again.
The FHEM GUI still doesn't update as expected...

See:

But homekit still doesn't update correctly. The current temperature is not updated too.

So I guess there must be some sort of bug in homebridge-fhem... It's really hard to identify the source of the problem...

from homebridge-fhem.

you say the events are visible in the event monitor but fhemweb itself will not update the readings in the device detail view?

is the event visible in the homebridge console output?

from homebridge-fhem.

you say the events are visible in the event monitor but fhemweb itself will not update the readings in the device detail view?

yes, it only updates when I refresh the browser window. Other devices such as pilight or homematic do update their state in fhemweb automatically without problem.

is the event visible in the homebridge console output?

I just restarted homebrige with --debug flag and waited. The max thermostat events that are shown in the fhem event monitor are not shown in homebridge console output. There is no reaction at all. There only is the following message reappearing every ~20 seconds:

starting longpoll: http://127.0.0.1:8083/fhem.pl?XHR=1&inform=type=status;addglobal=1;filter=.*;since=null;fmt=JSON&timestamp=1478553387341

I guess that's normal.

For example; changing the state of a pilight switch results in console output in homebridge as expected.

from homebridge-fhem.

if fhemweb is also not updating automatically then it is not only a homebridge problem. but i have currently no idea what it is. please have a look at the javascript console of your browser. are there any error in the device detail page if such an event comes in?

longpoll should not be retarted every 20 seconds but a connection should last for hours or days.

there is something wrong...

from homebridge-fhem.

fhemweb seems to loose the connection on a regular basis too....

But there aren't any event related errors reported.
I've double checked my console filter; errors aren't hidden.

from homebridge-fhem.

Well, I do have a second fhem & homebridge instance on another raspberry pi.
The re-polling issue doesn't appear there (it doesn't have any Max! devices).

So I guess, there are two possible error sources:

• messed up configuration, so it's my fault
• broken Max! plug-in

In order to rule out the second option, I will remove the max_lan device from the broken instance and add it to the working instance. If the problem doesn't re-appear on the other instance, I have to assume it's my fault. So I will reinstall the whole raspbian (including FHEM and homebridge).

from homebridge-fhem.

After removing the max_lan device, the re-polling issue disappeared...

from homebridge-fhem.

could you please check if fhem is blocked before the reconnect happens?

from homebridge-fhem.

Do you mean the black & white message "Connection lost. Trying again in 5 seconds"?

from homebridge-fhem.

yes

from homebridge-fhem.

Yes it appears in most of the connection loss cases (re-polling messages on the js console)

from homebridge-fhem.

I think it's a max_lan (ml) issue. So I'm going to close this issue and I will try to use a CUL or CUNO instead.

from homebridge-fhem.