Feature Request: Watchdog

[taucher4000]

Hallo,

leider passiert es bei mir von Zeit-zu-Zeit, dass Node-Red aus noch unbekannten gründen abstürzt. Wäre es nicht super, wenn in RedMatic eine Art Watchdog laufen würde, welcher den NodeRed Dienst überwacht und bei bedarf neu startet?

Gruß

[Hypnos3]

Hier sollte man zuallererst die Ursache für den Absturz finden.

• steht was im logfile?
• hast du mehrere ccu-connection nodes?

Eine Art watchdog könnte man sich über ein ReGa Script zusammenbasteln.

[taucher4000]

Hi Hypnos3,

heute ist es schon wieder passiert. Die letzten einträge der Log-Datei sehen wie folgt aus:

Nov 30 17:44:09 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] publishEvent ["nr_9ee5_BidCos-RF","OEQ0238264:4","PARTY_START_TIME",0]
Nov 30 17:44:09 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] publishEvent ["nr_9ee5_BidCos-RF","OEQ0238264:4","PARTY_START_DAY",1]
Nov 30 17:44:09 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] publishEvent ["nr_9ee5_BidCos-RF","OEQ0238264:4","PARTY_START_MONTH",1]
Nov 30 17:44:09 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] publishEvent ["nr_9ee5_BidCos-RF","OEQ0238264:4","PARTY_START_YEAR",0]
Nov 30 17:44:09 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] publishEvent ["nr_9ee5_BidCos-RF","OEQ0238264:4","PARTY_STOP_TIME",0]
Nov 30 17:44:09 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] publishEvent ["nr_9ee5_BidCos-RF","OEQ0238264:4","PARTY_STOP_DAY",1]
Nov 30 17:44:09 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] publishEvent ["nr_9ee5_BidCos-RF","OEQ0238264:4","PARTY_STOP_MONTH",1]
Nov 30 17:44:09 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] publishEvent ["nr_9ee5_BidCos-RF","OEQ0238264:4","PARTY_STOP_YEAR",0]
Nov 30 17:44:09 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost]     > BidCos-RF system.multicall ["","","","","","","","","","","","","","",""]
Nov 30 17:44:19 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] rpcCheckInit BidCos-RF 9 60
Nov 30 17:44:22 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] getRegaVariables
Nov 30 17:44:22 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] getRegaPrograms
Nov 30 17:44:24 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] rpc < BidCos-RF system.multicall [[{"methodName":"event","params":["nr_9ee5_BidCos-RF","OEQ0612882:0","CONFIG_PENDING",
true]}]]
Nov 30 17:44:24 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] publishEvent ["nr_9ee5_BidCos-RF","OEQ0612882:0","CONFIG_PENDING",true]
Nov 30 17:44:24 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost]     > BidCos-RF system.multicall [""]
Nov 30 17:44:24 ccu3 daemon.debug node-red[15057]: [ccu-connection:localhost] rpc < BidCos-RF system.multicall [[{"methodName":"updateDevice","params":["nr_9ee5_BidCos-RF","OEQ0612882:4",1]}]]
Nov 30 17:44:24 ccu3 daemon.err node-red: 30 Nov 17:44:24 - [red] Uncaught Exception:
Nov 30 17:44:24 ccu3 daemon.err node-red: 30 Nov 17:44:24 - TypeError: params is not iterable
Nov 30 17:44:24 ccu3 daemon.err node-red:     at Object.updateDevice (/usr/local/addons/redmatic/var/node_modules/node-red-contrib-ccu/nodes/ccu-connection.js:1274:38)
Nov 30 17:44:24 ccu3 daemon.err node-red:     at params.(anonymous function).forEach.call (/usr/local/addons/redmatic/var/node_modules/node-red-contrib-ccu/nodes/ccu-connection.js:1381:61)
Nov 30 17:44:24 ccu3 daemon.err node-red:     at Array.forEach (<anonymous>)
Nov 30 17:44:24 ccu3 daemon.err node-red:     at Object.system.multicall (/usr/local/addons/redmatic/var/node_modules/node-red-contrib-ccu/nodes/ccu-connection.js:1354:31)
Nov 30 17:44:24 ccu3 daemon.err node-red:     at Server.servers.(anonymous function).on (/usr/local/addons/redmatic/var/node_modules/node-red-contrib-ccu/nodes/ccu-connection.js:1076:48)
Nov 30 17:44:24 ccu3 daemon.err node-red:     at emitThree (events.js:136:13)
Nov 30 17:44:24 ccu3 daemon.err node-red:     at Server.emit (events.js:217:7)
Nov 30 17:44:24 ccu3 daemon.err node-red:     at Server.handleCall (/usr/local/addons/redmatic/var/node_modules/node-red-contrib-ccu/node_modules/binrpc/lib/server.js:79:24)
Nov 30 17:44:24 ccu3 daemon.err node-red:     at Socket.<anonymous> (/usr/local/addons/redmatic/var/node_modules/node-red-contrib-ccu/node_modules/binrpc/lib/server.js:58:22)
Nov 30 17:44:24 ccu3 daemon.err node-red:     at emitOne (events.js:116:13)

Leider kann ich dem nicht viel entnehmen. Du evtl.? Ich habe nur eine CCU-Connection node aktiv.

[Hypnos3]

Ja,
den gleichen Fehler hatte ich auch und dieser ist in der aktuellen Version behoben.

Entweder du aktualisierst RedMatic über die WebUI komplett oder nur node-red-contrib-ccu.

Um nur node-red-contrib-ccu zu aktualisieren folgendes vorgehen:

danach auf update

wenn das fertig ist musst du jedoch Node-Red neu starten.

[hobbyquaker]

Hab nun einen automatischen Neustart von Node-RED im Falle eines Absturzes eingebaut. Aber dieses Feature bitte nicht missverstehen ;) Natürlich darf Node-RED nicht abstürzen und wenn es das tut müssen wir die Ursache finden und beheben. In der RedMatic UI (CCU WebUI Systemeinstellungen -> RedMatic) wird nun angezeigt wie oft so ein automatischer Neustart stattgefunden hat.

[Hypnos3]

Gibt es da ein Limit? Ich stell mir gerade eine Endlosschleife vor. Ich hatte schon mal einen kaputten Flow, der Node-Red beim starten zum Asturz gebracht hatte. Es sollte höchstens ein neustart innerhalb einer halben Stunde stattfinden oder was Grundlegenderes ist kaputt.

[psi-4ward]

node-red restart hat ja auch einfluss auf de dutycycle. das könnte böse überraschungen geben.

wenigestens nen sleep 5 rein oder so

[hobbyquaker]

mja, da stimm ich euch zu, war vielleicht keine so spitzen idee das ohne limits einzubauen. ich werde es mal auf eine konfigurierbare anzahl begrenzbar und abschaltbar machen, besser ist das. 5 Sekunden sleep bringts glaub ich nicht so, das Node-RED braucht eh ziemlich lang für den neustart und wenn jemand Flows hat die beim start den Dutycycle hochtreiben würde man das eher in der Größenordnung 15-30 Minuten mitigaten. Aber von solchen Flows (die viele setValue beim Start durchführen) ist auch abzuraten.

[hobbyquaker]

noch ne idee: man sollte beim Einsatz dieses features dazu raten einen flow zu bauen der beim start ne Pushover/Email oder so sendet, damit man es wenigstens mitbekommt dass da was schiefläuft.

[Hypnos3]

Ja, sowas hab ich am Start.

Hier gleich ein beispiel:

[{"id":"cab4b267.9480d","type":"delay","z":"a258f3d1.6da17","name":"","pauseType":"delay","timeout":"5","timeoutUnits":"seconds","rate":"1","nbRateUnits":"1","rateUnits":"second","randomFirst":"1","randomLast":"5","randomUnits":"seconds","drop":false,"x":780,"y":480,"wires":[["66b982b.b728d7c","1dd1cb9a.84d4a4"]]},{"id":"ba1ca2de.fc4bf","type":"comment","z":"a258f3d1.6da17","name":"Systemfunktionen - Start","info":"","x":140,"y":40,"wires":[]},{"id":"66b982b.b728d7c","type":"change","z":"a258f3d1.6da17","name":"RUNLEVEL=3 / RUNNING=true","rules":[{"t":"set","p":"#:(memory)::RUNLEVEL","pt":"global","to":"3","tot":"num"},{"t":"set","p":"#:(memory)::RUNNING","pt":"global","to":"true","tot":"bool"},{"t":"set","p":"payload","pt":"msg","to":"3","tot":"num"}],"action":"","property":"","from":"","to":"","reg":false,"x":1070,"y":480,"wires":[["e827db51.0ab838","d3543900.4d1718"]]},{"id":"fc1672a0.1b6ed","type":"change","z":"a258f3d1.6da17","name":"RUNNING=false / RUNLEVEL=0","rules":[{"t":"set","p":"#:(memory)::RUNNING","pt":"global","to":"false","tot":"bool"},{"t":"set","p":"#:(memory)::RUNLEVEL","pt":"global","to":"0","tot":"num"}],"action":"","property":"","from":"","to":"","reg":false,"x":450,"y":100,"wires":[["155bb4d1.2558cb"]]},{"id":"294b6397.84943c","type":"link out","z":"a258f3d1.6da17","name":"chg-Startup-RunLevel1","links":["d9fbc18e.7df69"],"x":1275,"y":220,"wires":[]},{"id":"e827db51.0ab838","type":"link out","z":"a258f3d1.6da17","name":"chg-Startup-RunLevel3","links":["c2700f38.8baf2","2509748.1503e8c","334a5a27.1c55b6","cfc5366c.37c388"],"x":1275,"y":480,"wires":[]},{"id":"5e8e8049.525a7","type":"delay","z":"a258f3d1.6da17","name":"","pauseType":"delay","timeout":"5","timeoutUnits":"seconds","rate":"1","nbRateUnits":"1","rateUnits":"second","randomFirst":"1","randomLast":"5","randomUnits":"seconds","drop":false,"x":780,"y":420,"wires":[["cab4b267.9480d","be39b899.534578"]]},{"id":"39805c97.f8e0d4","type":"link out","z":"a258f3d1.6da17","name":"chg-Startup-RunLevel2","links":["8158c97c.7c9478"],"x":1275,"y":420,"wires":[]},{"id":"b9336030.cdf32","type":"change","z":"a258f3d1.6da17","name":"RUNLEVEL=1","rules":[{"t":"set","p":"#:(memory)::RUNLEVEL","pt":"global","to":"1","tot":"num"},{"t":"set","p":"payload","pt":"msg","to":"1","tot":"num"}],"action":"","property":"","from":"","to":"","reg":false,"x":1020,"y":220,"wires":[["5e8e8049.525a7","294b6397.84943c","d3543900.4d1718","197916ec.24a339"]]},{"id":"5f4fac47.824ec4","type":"comment","z":"a258f3d1.6da17","name":"chg-Startup-RunLevel1","info":"","x":1520,"y":220,"wires":[]},{"id":"1c2fad16.462943","type":"comment","z":"a258f3d1.6da17","name":"chg-Startup-RunLevel2","info":"","x":1540,"y":420,"wires":[]},{"id":"6a12612e.1aec8","type":"comment","z":"a258f3d1.6da17","name":"chg-Startup-RunLevel3","info":"","x":1540,"y":480,"wires":[]},{"id":"be39b899.534578","type":"change","z":"a258f3d1.6da17","name":"RUNLEVEL = 2","rules":[{"t":"set","p":"#:(memory)::RUNLEVEL","pt":"global","to":"2","tot":"num"},{"t":"set","p":"payload","pt":"msg","to":"2","tot":"num"}],"action":"","property":"","from":"","to":"","reg":false,"x":1020,"y":420,"wires":[["39805c97.f8e0d4","d3543900.4d1718"]]},{"id":"d3543900.4d1718","type":"debug","z":"a258f3d1.6da17","name":"System Start","active":true,"tosidebar":false,"console":true,"tostatus":false,"complete":"payload","x":1340,"y":360,"wires":[]},{"id":"bb7a8bcb.11e758","type":"inject","z":"a258f3d1.6da17","name":"","topic":"startup","payload":"start","payloadType":"str","repeat":"","crontab":"","once":true,"onceDelay":"0.001","x":140,"y":100,"wires":[["fc1672a0.1b6ed"]]},{"id":"d618593f.dd0b28","type":"change","z":"a258f3d1.6da17","name":"eMail Message","rules":[{"t":"set","p":"subject","pt":"msg","to":"Restart","tot":"str"},{"t":"set","p":"payload","pt":"msg","to":"\"Neustart der Zentrale (Node-Red)!! \" & $$.payload","tot":"jsonata"}],"action":"","property":"","from":"","to":"","reg":false,"x":1580,"y":280,"wires":[["dc7ea158.18c0d"]]},{"id":"155bb4d1.2558cb","type":"unsafe-function","z":"a258f3d1.6da17","name":"check Date","func":"if (Date.now() < 1500000000000 ) {\n    return [null, msg];\n}\nreturn msg;","outputs":2,"noerr":0,"x":530,"y":220,"wires":[["47804b32.089464"],["1acd1392.8b493c"]]},{"id":"1acd1392.8b493c","type":"delay","z":"a258f3d1.6da17","name":"","pauseType":"random","timeout":"8","timeoutUnits":"seconds","rate":"1","nbRateUnits":"1","rateUnits":"second","randomFirst":"5","randomLast":"12","randomUnits":"seconds","drop":false,"x":520,"y":300,"wires":[["155bb4d1.2558cb"]]},{"id":"47804b32.089464","type":"delay","z":"a258f3d1.6da17","name":"","pauseType":"delay","timeout":"500","timeoutUnits":"milliseconds","rate":"1","nbRateUnits":"1","rateUnits":"second","randomFirst":"1","randomLast":"5","randomUnits":"seconds","drop":false,"x":770,"y":220,"wires":[["b9336030.cdf32"]]},{"id":"1dd1cb9a.84d4a4","type":"delay","z":"a258f3d1.6da17","name":"","pauseType":"delay","timeout":"10","timeoutUnits":"seconds","rate":"1","nbRateUnits":"1","rateUnits":"second","randomFirst":"1","randomLast":"5","randomUnits":"seconds","drop":false,"x":780,"y":540,"wires":[["bacca792.36af38"]]},{"id":"8dc03e30.6fa58","type":"link out","z":"a258f3d1.6da17","name":"chg-Startup-RunLevel4","links":["9ceac20c.33d19","f98c6394.b06e1"],"x":1275,"y":540,"wires":[]},{"id":"bacca792.36af38","type":"change","z":"a258f3d1.6da17","name":"RUNLEVEL=4","rules":[{"t":"set","p":"#:(memory)::RUNLEVEL","pt":"global","to":"4","tot":"num"},{"t":"set","p":"payload","pt":"msg","to":"4","tot":"num"}],"action":"","property":"","from":"","to":"","reg":false,"x":1020,"y":540,"wires":[["8dc03e30.6fa58","d3543900.4d1718"]]},{"id":"1b30d1d3.3b1cce","type":"comment","z":"a258f3d1.6da17","name":"chg-Startup-RunLevel4","info":"","x":1540,"y":540,"wires":[]},{"id":"197916ec.24a339","type":"exec","z":"a258f3d1.6da17","command":"cat /boot/VERSION | cut -d'=' -f2","addpay":false,"append":"","useSpawn":"false","timer":"5","oldrc":false,"name":"get firmware","x":1330,"y":280,"wires":[["d618593f.dd0b28"],["d618593f.dd0b28"],[]]},{"id":"dc7ea158.18c0d","type":"e-mail","z":"a258f3d1.6da17","server":"smtp.1und1.de","port":"465","secure":true,"name":"","dname":"","x":1770,"y":280,"wires":[]}]

kann gleich mit in die Doku.

[Hypnos3]

schickt beim Start die aktuelle firmware gleich mit und wartet, wenn der ntp noch nicht bereit ist.

[psi-4ward]

zwar klever aber geht schon bissi weit für den otto normal user?
vllt ne Servicemeldung in der CCU erzeugen?