00_TCM DevIoJustClosed 1

Begonnen von Tomk, 06 März 2023, 22:10:29

Vorheriges Thema - Nächstes Thema

Tomk

Hallo, ich kämpfe seit Tagen damit meine Enocean Anbindung wieder ans laufen zu bekommen. Nach einem missglückten Update Versuch meines Jessie pi's habe ich ein SD Card Backup eingespielt. Danach lief mein enocean pi Modul nicht mehr. Habe mir jetzt ein usb 310 Stick bestellt. Hiermit habe ich jetzt folgendes Problem:

Nach erfolgreicher Initialisierung konnte ich die base id abfragen. Kurz darauf geht die Verbindung verloren.

Was mich wundert ist das bei den internals das hier steht: DevIoJustClosed 1. was hat das zu bedeuten?

2023.03.06 20:48:24 2: TCM TCM_ESP3 smartAckMailboxMax 0 restored
2023.03.06 20:48:24 2: TCM TCM_ESP3 initialized
2023.03.06 20:48:29 3: TCM TCM_ESP3 get baseID
2023.03.06 20:48:29 5: TCM TCM_ESP3 sent ESP: 5500010005700838
2023.03.06 20:48:29 5: DevIo_SimpleWrite TCM_ESP3: 5500010005700838
2023.03.06 20:48:29 2: TCM TCM_ESP3 No FD
2023.03.06 20:48:37 3: TCM TCM_ESP3 get getFreeID
2023.03.06 20:48:37 1: PERL WARNING: Use of uninitialized value in string ne at ./FHEM/10_EnOcean.pm line 16780.
2023.03.06 20:48:52 3: TCM TCM_ESP3 get frequencyInfo
2023.03.06 20:48:52 5: TCM TCM_ESP3 sent ESP: 55000100057025FB
2023.03.06 20:48:52 5: DevIo_SimpleWrite TCM_ESP3: 55000100057025FB
2023.03.06 20:48:52 2: TCM TCM_ESP3 No FD
2023.03.06 20:48:56 3: TCM TCM_ESP3 get filter
2023.03.06 20:48:56 5: TCM TCM_ESP3 sent ESP: 5500010005700F2D
2023.03.06 20:48:56 5: DevIo_SimpleWrite TCM_ESP3: 5500010005700F2D
2023.03.06 20:48:56 2: TCM TCM_ESP3 No FD


Das Aufsteckmodul habe ich inzwischen entfernt und alle usb devices sind über "by-Id" eingebunden...

Hat jemand einen Tipp?

Flachzange

Bitte mal das TCM device komplett löschen. Fhem neustarten und dann komplettes Log seit FHEM-Start und ein list vom TCM device anhängen.

olivierbardot

Hello !

I have exactly the same problem with my EnOcean TCM 310 module

First I have successfully re included the module after having it deleted.
The connexion is lost between a few hours and a few days after FHEM restart.
During this time, the module is fully working (included smartack on sensors).

A few things :
- this module has worked a few years without any problems before fhem and system update
- the module is on usb1 and RFXcom on usb0. RFXCom is not affected by any usb connexion problem
- TCM module (and RFX COM) are powered by usb hub with external power line. Pi with separate 2.5 A power supply.
- I have monitored pi CPU temperature and throttle and Enocean disconnections together (see enclosed file).

I have suspected a hardware or software problem on my raspberrypi 3B+ board but the "very strange thing" is that : FHEM restart re establish successfully the usb connection with the EnOcean TCM 310 which seemed to be simply "freezed".
A few hours after, the connection with the TCM 310 is lost (freezed) again.

I plan to upgrade the pi to raspbian bullseye in order to see if it solves the problem.

What is your advice for further investigations before installing raspbian bullseye on my pi ?

Thanks by advance
Olivier


Raspis EnOcean TrXCom Linky 1W Telegram Motion EgiGeozone OpenDTU HiveMQ

olivierbardot

Further, considering : https://forum.fhem.de/index.php/topic,130427.msg1246767.html#msg1246767,

I tried :
/sys/bus/usb/devices/1-1.3/power/control : auto ->on
/sys/bus/usb/devices/1-1/power/control : auto -> on

I will tell the results of these

Raspis EnOcean TrXCom Linky 1W Telegram Motion EgiGeozone OpenDTU HiveMQ

olivierbardot

Hello

The problem seems to be definitely solved

What I have done :

1- udev rules and scripts for :
/sys/bus/usb/devices/1-1.3/power/control
on
/sys/bus/usb/devices/1-1.3/power/level
on

with 1-1.3 : usb hub on which EnOcean devices are connected

2- raspi system update from Stretch to Buster

Hope this can help ...
Raspis EnOcean TrXCom Linky 1W Telegram Motion EgiGeozone OpenDTU HiveMQ