I installed a new iblind this evening and it is appearing in ZWaveJS. It is operational thru Home Assistant just fine. It is operational thru the HA ZwaveJS plugin just fine.
Where it is non-responsive is in MSR for some reason. The entities are there. Adding the node to a Reaction and then attempting to run said Reaction nets me this:
[latest-23242]2023-09-08T05:33:08.454Z <ZWaveJSController:5:ZWaveJSController.js:655> ZWaveJSController#zwavejs _apply_value motion_sensor.state=false [latest-23242]2023-09-08T05:33:08.455Z <ZWaveJSController:5:ZWaveJSController.js:722> ZWaveJSController#zwavejs setting Binary Sensor#zwavejs>11-0.x_zwave_values.Binary_Sensor_Motion to false [latest-23242]2023-09-08T05:33:08.692Z <ZWaveJSController:5:ZWaveJSController.js:360> ZWaveJSController#zwavejs handling node event statistics updated entity Binary Sensor#zwavejs>11-0 [latest-23242]2023-09-08T05:40:23.960Z <ZWaveJSController:INFO> ZWaveJSController#zwavejs performing cover.open on Cover#zwavejs>20-0 with [Object]{ } [latest-23242]2023-09-08T05:40:23.962Z <ZWaveJSController:5:ZWaveJSController.js:1843> ZWaveJSController#zwavejs no implementation mapped; attempting default [latest-23242]2023-09-08T05:40:25.062Z <ZWaveJSController:INFO> ZWaveJSController#zwavejs performing zwave_device.refresh on Cover#zwavejs>20-0 with [Object]{ } **[latest-23242]2023-09-08T05:40:25.063Z <ZWaveJSController:5:ZWaveJSController.js:1843> ZWaveJSController#zwavejs no implementation mapped; attempting default** [latest-23242]2023-09-08T05:40:25.065Z <ZWaveJSController:5:ZWaveJSController.js:294> ZWaveJSController#zwavejs sending #1694151625064<9/8/2023, 1:40:25 AM>: [Object]{ "command": "node.refresh_values", "nodeId": 20, "messageId": 1694151625064 } [latest-23242]2023-09-08T05:40:26.307Z <ZWaveJSController:5:ZWaveJSController.js:360> ZWaveJSController#zwavejs handling node event statistics updated entity Cover#zwavejs>20-0 [latest-23242]2023-09-08T05:40:26.317Z <ZWaveJSController:5:ZWaveJSController.js:360> ZWaveJSController#zwavejs handling node event value updated entity Cover#zwavejs>20-0 [latest-23242]2023-09-08T05:40:26.318Z <ZWaveJSController:5:ZWaveJSController.js:667> ZWaveJSController#zwavejs update node 20 value "0:128:level:" data [Object]{ "source": "node", "event": "value updated", "nodeId": 20, "args": { "commandClassName": "Battery", "commandClass": 128, "property": "level", "endpoint": 0, "newValue": 100, "prevValue": 100, "propertyName": "level" } } [latest-23242]2023-09-08T05:40:26.319Z <ZWaveJSController:5:ZWaveJSController.js:684> ZWaveJSController#zwavejs updating attributes for node 20 value "0:128:level:"=100: [Array][ "battery_power.level", "battery_power.since" ] [latest-23242]2023-09-08T05:40:26.320Z <ZWaveJSController:5:ZWaveJSController.js:698> ZWaveJSController#zwavejs updating attribute battery_power.level with [Object]{ "entity": "20-0", "impl": { "expr": "float( value ) / 100", "valueId": "128:level:" } } [latest-23242]2023-09-08T05:40:26.321Z <ZWaveJSController:5:ZWaveJSController.js:591> ZWaveJSController#zwavejs _apply_value entity Cover#zwavejs>20-0 battery_power.level [latest-23242]2023-09-08T05:40:26.322Z <ZWaveJSController:5:ZWaveJSController.js:655> ZWaveJSController#zwavejs _apply_value battery_power.level=1 [latest-23242]2023-09-08T05:40:26.323Z <ZWaveJSController:5:ZWaveJSController.js:698> ZWaveJSController#zwavejs updating attribute battery_power.since with [Object]{ "entity": "20-0", "impl": { "expr": "time()", "valueId": "128:level:" } } [latest-23242]2023-09-08T05:40:26.323Z <ZWaveJSController:5:ZWaveJSController.js:591> ZWaveJSController#zwavejs _apply_value entity Cover#zwavejs>20-0 battery_power.since [latest-23242]2023-09-08T05:40:26.324Z <ZWaveJSController:5:ZWaveJSController.js:655> ZWaveJSController#zwavejs _apply_value battery_power.since=1694151626324<9/8/2023, 1:40:26 AM> [latest-23242]2023-09-08T05:40:26.325Z <ZWaveJSController:5:ZWaveJSController.js:722> ZWaveJSController#zwavejs setting Cover#zwavejs>20-0.x_zwave_values.Battery_level to 100 [latest-23242]2023-09-08T05:40:26.333Z <ZWaveJSController:5:ZWaveJSController.js:360> ZWaveJSController#zwavejs handling node event value updated entity Cover#zwavejs>20-0 [latest-23242]2023-09-08T05:40:26.334Z <ZWaveJSController:5:ZWaveJSController.js:667> ZWaveJSController#zwavejs update node 20 value "0:128:isLow:" data [Object]{ "source": "node", "event": "value updated", "nodeId": 20, "args": { "commandClassName": "Battery", "commandClass": 128, "property": "isLow", "endpoint": 0, "newValue": false, "prevValue": false, "propertyName": "isLow" } } [latest-23242]2023-09-08T05:40:26.336Z <ZWaveJSController:5:ZWaveJSController.js:324> ZWaveJSController#zwavejs request 1694151625064<9/8/2023, 1:40:25 AM> (node.refresh_values) success notification [latest-23242]2023-09-08T05:40:26.337Z <ZWaveJSController:5:ZWaveJSController.js:684> ZWaveJSController#zwavejs updating attributes for node 20 value "0:128:isLow:"=false: [Array][ ] [latest-23242]2023-09-08T05:40:26.338Z <ZWaveJSController:5:ZWaveJSController.js:722> ZWaveJSController#zwavejs setting Cover#zwavejs>20-0.x_zwave_values.Battery_isLow to false [latest-23242]2023-09-08T05:40:26.558Z <ZWaveJSController:5:ZWaveJSController.js:360> ZWaveJSController#zwavejs handling node event statistics updated entity Cover#zwavejs>20-0Please note the highlighted logpart - no other iblind has this. In addition, in MSR>Entities several show as null rather than having a value as the others do.
69631fbc-db4e-4e8d-aa98-10565811cd93-image.png
I tried deleting all Entities associated with this device from MSR and then refreshing ZWaveJS in Home Assistant and restarting MSR - the Entities return but in the same state/s.
Admittedly, it's been a very long three weeks at work - I could have missed something obvious during setup but I sure can't figure what it is.
Hi @toggledbits
I'm trying to use OWM, apparently, I receive the correct information, but after a while, the MSR disconnects.
My configuration is as follows.
- id: weather enabled: true implementation: OWMWeatherController name: OWM Weather config: # Place your OWM API key here (remember to enable the controller after adding your appid) appid: "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" # How often weather is allowed to be refreshed. This helps limit OWN API use, # to keep you in their good graces (and on their free tier). This value is # in minutes. interval: 30 locations: - id: home name: Home Weather # Set the location by specifying ONE OF: latitude+longitude, OWN city # ID, or location (as postal,country). If none is set, the system # location will be used. latitude: 8.9936 longitude: -79.5197 city_id: 3703443 #location: "30269,us" # Enable "save_full_response" to save the full response (you may have # need to use parts of the response that are not part of the wx # capability in your dashboard widgets). #save_full_response: falseWhen I request a systemctl status reactor I get this message:
root@main:/home/wilson/reactor/logs# systemctl status reactor ● reactor.service - Multi System Reactor Loaded: loaded (/etc/systemd/system/reactor.service; enabled; vendor preset: enabled) Active: active (running) since Fri 2023-09-08 21:03:13 EST; 5s ago Main PID: 194711 (node) Tasks: 11 (limit: 9390) Memory: 82.5M CPU: 2.508s CGroup: /system.slice/reactor.service └─194711 /usr/bin/node app -p Sep 08 21:03:13 main node[194711]: Error: Incompatible serialization data; can't unserialize Sep 08 21:03:13 main node[194711]: at System.unserialize (/home/wilson/reactor/server/lib/Entity.js:624:19) Sep 08 21:03:13 main node[194711]: at /home/wilson/reactor/server/lib/Controller.js:458:70 Sep 08 21:03:13 main node[194711]: at Array.forEach (<anonymous>) Sep 08 21:03:13 main node[194711]: at OWMWeatherController._restoreEntities (/home/wilson/reactor/server/lib/Controller.js:446:36) Sep 08 21:03:13 main node[194711]: at new Controller (/home/wilson/reactor/server/lib/Controller.js:37:45) Sep 08 21:03:13 main node[194711]: at new OWMWeatherController (/home/wilson/reactor/server/lib/OWMWeatherController.js:327:9) Sep 08 21:03:13 main node[194711]: at /home/wilson/reactor/server/lib/Controller.js:93:37 Sep 08 21:03:13 main node[194711]: [latest-23242]2023-09-09T02:03:13.855Z <NUTController:null> Module NUTController v22305 Sep 08 21:03:13 main node[194711]: [latest-23242]2023-09-09T02:03:13.859Z <SystemController:null> Module SystemController v23214 root@main:/home/wilson/reactor/logs# root@main:/home/wilson/reactor# systemctl status reactor ● reactor.service - Multi System Reactor Loaded: loaded (/etc/systemd/system/reactor.service; enabled; vendor preset: enabled) Active: active (running) since Fri 2023-09-08 20:48:36 EST; 2min 58s ago Main PID: 194353 (node) Tasks: 11 (limit: 9390) Memory: 66.0M CPU: 9.493s CGroup: /system.slice/reactor.service └─194353 /usr/bin/node app -p Sep 08 20:48:37 main node[194353]: at System.unserialize (/home/wilson/reactor/server/lib/Entity.js:624:19) Sep 08 20:48:37 main node[194353]: at /home/wilson/reactor/server/lib/Controller.js:458:70 Sep 08 20:48:37 main node[194353]: at Array.forEach (<anonymous>) Sep 08 20:48:37 main node[194353]: at OWMWeatherController._restoreEntities (/home/wilson/reactor/server/lib/Controller.js:446:36) Sep 08 20:48:37 main node[194353]: at new Controller (/home/wilson/reactor/server/lib/Controller.js:37:45) Sep 08 20:48:37 main node[194353]: at new OWMWeatherController (/home/wilson/reactor/server/lib/OWMWeatherController.js:327:9) Sep 08 20:48:37 main node[194353]: at /home/wilson/reactor/server/lib/Controller.js:93:37 Sep 08 20:48:37 main node[194353]: [latest-23242]2023-09-09T01:48:37.017Z <NUTController:null> Module NUTController v22305 Sep 08 20:48:37 main node[194353]: [latest-23242]2023-09-09T01:48:37.024Z <SystemController:null> Module SystemController v23214 Sep 08 20:48:37 main node[194353]: [latest-23242]2023-09-09T01:48:37.475Z <Timer:null> Timer#rule-l7ujwva5 just a note: I'm setting a delay of > lines 1-20/20 (END)I can see this information in the log, but I don't understand what I should do to fix it.
[latest-23242]2023-09-09T02:03:13.838Z <DynamicGroupController:null> Module DynamicGroupController v22313 [latest-23242]2023-09-09T02:03:13.845Z <Structure:INFO> Structure#1 loading controller interface weather (OWMWeatherController) [latest-23242]2023-09-09T02:03:13.847Z <OWMWeatherController:null> Module OWMWeatherController v22294 [latest-23242]2023-09-09T02:03:13.850Z <Controller:WARN> OWMWeatherController#weather failed (1) to restore entity controller_all: [Error] Inco> [latest-23242]2023-09-09T02:03:13.850Z <Controller:CRIT> Error: Incompatible serialization data; can't unserialize [-] Error: Incompatible serialization data; can't unserialize at Group.unserialize (/home/wilson/reactor/server/lib/Entity.js:624:19) at /home/wilson/reactor/server/lib/Controller.js:458:70 at Array.forEach (<anonymous>) at OWMWeatherController._restoreEntities (/home/wilson/reactor/server/lib/Controller.js:446:36) at new Controller (/home/wilson/reactor/server/lib/Controller.js:37:45) at new OWMWeatherController (/home/wilson/reactor/server/lib/OWMWeatherController.js:327:9) at /home/wilson/reactor/server/lib/Controller.js:93:37 [latest-23242]2023-09-09T02:03:13.851Z <Controller:WARN> OWMWeatherController#weather failed (1) to restore entity default: [Error] Incompatibl> [latest-23242]2023-09-09T02:03:13.851Z <Controller:CRIT> Error: Incompatible serialization data; can't unserialize [-] Error: Incompatible serialization data; can't unserialize at Entity.unserialize (/home/wilson/reactor/server/lib/Entity.js:624:19) at /home/wilson/reactor/server/lib/Controller.js:458:70 at Array.forEach (<anonymous>) at OWMWeatherController._restoreEntities (/home/wilson/reactor/server/lib/Controller.js:446:36) at new Controller (/home/wilson/reactor/server/lib/Controller.js:37:45) at new OWMWeatherController (/home/wilson/reactor/server/lib/OWMWeatherController.js:327:9) at /home/wilson/reactor/server/lib/Controller.js:93:37 [latest-23242]2023-09-09T02:03:13.852Z <Controller:WARN> OWMWeatherController#weather failed (1) to restore entity system: [Error] Incompatible> [latest-23242]2023-09-09T02:03:13.852Z <Controller:CRIT> Error: Incompatible serialization data; can't unserialize [-] Error: Incompatible serialization data; can't unserialize at System.unserialize (/home/wilson/reactor/server/lib/Entity.js:624:19) at /home/wilson/reactor/server/lib/Controller.js:458:70 at Array.forEach (<anonymous>) at OWMWeatherController._restoreEntities (/home/wilson/reactor/server/lib/Controller.js:446:36) at new Controller (/home/wilson/reactor/server/lib/Controller.js:37:45) at new OWMWeatherController (/home/wilson/reactor/server/lib/OWMWeatherController.js:327:9) at /home/wilson/reactor/server/lib/Controller.js:93:37 [latest-23242]2023-09-09T02:03:13.853Z <Structure:INFO> Structure#1 loading controller interface nut (NUTController) [latest-23242]2023-09-09T02:03:13.855Z <NUTController:null> Module NUTController v22305 [latest-23242]2023-09-09T02:03:13.855Z <Controller:INFO> Loaded NUTController version "0.1.22305"; Patrick Rigney/Kedron Holdings LLC <patrick@> [latest-23242]2023-09-09T02:03:13.857Z <Structure:INFO> Structure#1 loading controller interface reactor_system (SystemController) [And finally, I have a second house, with the same OWM configuration I made today, and it doesn't show any errors.
Please, your traditional help.
Thanks.
After a prolonged power cut, the UPS was unable to cope with the weather and shut down the entire system.
When I returned, I had the problem that the host time displayed in the browser and automatically used in the rules was not correct.
I searched for a topic here on this forum that talked about it, but it referred to a container installation. There was a recommendation to go into the storage/states directory and delete the files and restart MSR, I did it and it didn't solve the problem.
There was also a recommendation to go directly to the browser to check the time, and I did so and I see that the time is correct.
8ad429c8-ea5f-45b4-8670-7d0353a50a0e-image.png
As you can see on the screen above:
MSR browser shows the correct browser time 21:23, but host time 13:02; In the browser below the direct query and the host time is correct 9:22:54 PM; Setting the VM date command on the third screen, the time is 9:23:00 PM.I've opened the entire reactor.log file and searched for any errors, and there aren't any.
^L [latest-23242]2023-09-15T02:11:12.928Z <app:null> Reactor build latest-23242-5ee8e1d4 starting on v18.12.0 [latest-23242]2023-09-15T02:11:12.929Z <app:null> Process ID 1807 user/group 1000/1000; bare-metal; platform linux/x64 #179-Ubuntu SMP Mo> [latest-23242]2023-09-15T02:11:12.929Z <app:null> Basedir /home/leonardo/reactor; data in /home/leonardo/reactor/storage [latest-23242]2023-09-15T02:11:12.929Z <app:null> NODE_PATH=/home/leonardo/reactor [latest-23242]2023-09-15T02:11:12.933Z <app:INFO> Configured locale (undefined); selected locale(s) en_US.UTF-8 [latest-23242]2023-09-15T02:11:12.955Z <app:INFO> Loaded locale en-US for en-US [latest-23242]2023-09-15T02:11:12.958Z <Structure:null> Module Structure v23172 [latest-23242]2023-09-15T02:11:12.959Z <Capabilities:null> Module Capabilities v22356 [latest-23242]2023-09-15T02:11:12.969Z <Capabilities:NOTICE> System capabilities loaded from core distribution, data version 23914 revisi> [latest-23242]2023-09-15T02:11:12.978Z <Plugin:null> Module Plugin v22300 [latest-23242]2023-09-15T02:11:12.981Z <TimerBroker:null> Module TimerBroker v22283 [latest-23242]2023-09-15T02:11:12.983Z <Entity:null> Module Entity v22353 [latest-23242]2023-09-15T02:11:12.986Z <Controller:null> Module Controller v23069 [latest-23242]2023-09-15T02:11:12.992Z <default:null> Module Ruleset v22293 [latest-23242]2023-09-15T02:11:12.992Z <default:null> Module Rulesets v22146 [latest-23242]2023-09-15T02:11:12.996Z <GlobalExpression:null> Module GlobalExpression v23211 [latest-23242]2023-09-15T02:11:13.003Z <Predicate:null> Module Predicate v23093 [latest-23242]2023-09-15T02:11:13.005Z <AlertManager:null> Module AlertManager v22283 [latest-23242]2023-09-15T02:11:13.007Z <Rule:null> Module Rule v23107 [latest-23242]2023-09-15T02:11:13.009Z <GlobalReaction:null> Module GlobalReaction v22324 [latest-23242]2023-09-15T02:11:13.010Z <Engine:null> Module Engine v23231 [latest-23242]2023-09-15T02:11:13.012Z <httpapi:null> Module httpapi v23058 [latest-23242]2023-09-15T02:11:13.019Z <wsapi:null> Module wsapi v23172 [latest-23242]2023-09-15T02:11:13.019Z <app:NOTICE> Starting Structure...As you can see, the time in the log is 02:11, which is correct, because I'm in time zone -5, which means it's 09:11PM for me.
I've restarted the VM, the computer hosting the VM, and the MSR a few times and the same difference remains. What do you recommend for the next step in understanding and solving the problem?
Thanks.
Build 21228 has been released. Docker images available from DockerHub as usual, and bare-metal packages here.
Home Assistant up to version 2021.8.6 supported; the online version of the manual will now state the current supported versions; Fix an error in OWMWeatherController that could cause it to stop updating; Unify the approach to entity filtering on all hub interface classes (controllers); this works for device entities only; it may be extended to other entities later; Improve error detail in messages for EzloController during auth phase; Add isRuleSet() and isRuleEnabled() functions to expressions extensions; Implement set action for lock and passage capabilities (makes them more easily scriptable in some cases); Fix a place in the UI where 24-hour time was not being displayed.Good afternoon,
I'm running my MSR instance on an old NAS server running Fedora 37 Server. This is an older Core i5-4690K server that has served me well overall, but it's showing it's age, and will often reboot into maint mode if it applies updates overnight. I've been trying to troubleshoot, but it's also in my office (makes my office hot), and I'm looking at getting either a QNAP or Syslogy rack mount NAS to replace it.
What I'm wondering is what the process would be to migrate my current MSR config and move it to a different platform. I have quite a large ruleset that I don't want to have to recreate. I know I can run it in a container on either platform, but I'm not sure how I would migrate all my rules over. Is there one config file that has all of the rules, or could I simply copy my /reactor directory over to a enw platform?
I am also considering moving the MSR instance to a stand alone RPI 4 with a SSD. This way, while I'm migrating all of my files off of my NAS, I'm not loosing Reactor's functionality until I have whichever rack mount NAS I go with up and running, with all of the data migrated.
Thoughts?
Hello,
I've installed MSR in a Docker container on my Synology NAS. I'm able to access my Vera, but I'm having trouble using the Reactor import within the Docker environment. I'm using the Docker terminal, and I've navigated to "/opt/reactor". However, when I try to run the command, it's telling me I'm not in the installation directory, even though I can see the "tools" directory and the "apps.js" file there. I've also tried changing the file permissions, but nothing seems to work. How do you manage this in a Synology Docker container? Do you use the terminal inside Docker, or SSH into the Synology itself? I'm not finding the "/opt/reactor" path, but I see a "/var/reactor" directory, which I placed in my storage folder.
ERROR: This script must only be run from the Reactor app directory.
Maybe I'm missing something, but this appears to be a bug.
I'm troubleshooting a rule to see if one of the system participants is arriving (presence changes from not home to home). I want this arriving state to be less than 5 minutes (300 seconds), to differentiate between being home for a longer period of time, and just getting home.
In the restrictions sections of the Trigger, I set "Condition must be sustained for less than 300 seconds", and I save the trigger.
After I save and exit, I'm looking at the rule, and it's changed itself to x_hass.state == home; for at least 300 secs. I go back into edit the rule, and it in fact shows "at least" from the drop down.
This logic is of course failing and I can't seem to get it to work.
Changed to less than, and saved:
10a57ff9-4d8c-46b3-b594-399ef2a6d320-image.png
Exited:
a81587d6-f0fa-4518-ad00-a9f601f14509-image.png
Back in rule:
b8e27dd2-e6e7-4f93-844e-83e2b11e976d-image.png
I'm running latest-23218-ebab7a1b.
I added the Proximity integration to HA to create automations based on if a person is on vacation or not. The integration creates an entity and its x_hass.state is the distance to the zone you have setup in HA. For some reason MSR sees that value as a string so I am not able to use the<, >, etc operators as a trigger. Is there anyway this could be fixed without needing to convert the string in a local expression?
MSR 23218
HA 2023.8.3
Entity Attributes if needed
x_hass.domain="proximity" x_hass.entity_id="proximity.arman_proximity" x_hass.services=["proximity"] x_hass.state="0" x_hass_attr.dir_of_travel="unknown" x_hass_attr.friendly_name="arman_proximity" x_hass_attr.nearest="Arman’s iPhone" x_hass_attr.unit_of_measurement="km"MSR Version: latest-23218
Running as Docker Container : Docker version 24.0.5, build ced0996
OS: Linux 15BPMSR1 6.1.26-05272-g26c406245a2c #1 SMP PREEMPT_DYNAMIC Thu Apr 27 10:15:40 UTC 2023 aarch64 aarch64 aarch64 GNU/Linux
SMTP notifications had been working fine and I am unsure exactly what action(s) stopped it from working.
I have verified that the SMTP server configured in notifications.yml is working properly.
To duplicate the problem I am forcing an SMTP email notification using the set reaction play button, the reactor.log shows the following errors:
[latest-23218]2023-08-21T14:23:22.462Z <wsapi:ERR> [WSAPI]wsapi#1 can't queue reaction rule-lgxmrba4:S: [Error] Engine has not been started [-] [latest-23218]2023-08-21T14:23:22.463Z <wsapi:CRIT> Error: Engine has not been started [-] Error: Engine has not been started at Engine.queueReaction (/opt/reactor/server/lib/Engine.js:1494:46) at WSAPI.clientMessage (/opt/reactor/server/wsapi/wsapi.js:782:64) at WebSocket.<anonymous> (/opt/reactor/server/wsapi/wsapi.js:301:157) at WebSocket.emit (node:events:527:28) at WebSocket.emit (node:domain:475:12) at Receiver.receiverOnMessage (/opt/reactor/node_modules/ws/lib/websocket.js:1068:20) at Receiver.emit (node:events:527:28) at Receiver.emit (node:domain:475:12) at Receiver.dataMessage (/opt/reactor/node_modules/ws/lib/receiver.js:517:14) at Receiver.getData (/opt/reactor/node_modules/ws/lib/receiver.js:435:17)I have tried:
OS level: apt udate and apt upgrade and reboot
docker pull toggledbits/reactor:latest-aarch64 latest-aarch64: Pulling from toggledbits/reactor Digest: sha256:d8a019d457c9fef61f7c3e6908db1f1d0a2d14b6f7c5e9d4fc536c791503d152 Status: Image is up to date for toggledbits/reactor:latest-aarch64 docker.io/toggledbits/reactor:latest-aarch64docker stop reactor
docker start reactor
docker compose down
docker compose up -d
To no avail.
I feel like I'm missing something obvious but can't put my finger on it.
Thank you in advance for any guidance.
-bh
I have a rule that's driving my pool heater to match the solar excess. In the previous iteration, I checked for the status every minute, but I wanted something different and smarter. The logic is basically:
run continuously from 9:15 to 18:15 check for conditions, and if there's solar excess, use the power to turn on the heater, then wait 1 minute to check again turn off the heater, then wait 5 minutes to prevent frequent on/off cyclesI thought that this could be very similar to Reactor-Ex machina's logic, so I crafted a rule that's basically doing the same: an infinite loop from 9:15 to 18:15 that's checking for status and has a delay based on the heater's condition.
All went well, but after a couple of hours, I got this:
Reaction task "49683" failed due to an exception. Please see the log. rule-liy7z3gf:S (Pool-Heater-Solar) step 0: Error: Task has run too much! Runaway?and the rule was reset (and the heater was left on). Is that by design and I should adopt another strategy? What are others doing when they have a state machine?
With the EV charger I did two groups of condition, each triggering every x minutes based on the charger status, but this seems less smarter and very rudimental. Any suggestion is appreciated. Thanks.
I have a Tripp-Lite UPS (https://www.tripplite.com/smartpro-lcd-120v-1500va-900w-line-interactive-ups-avr-tower-lcd-usb-10-outlets~smart1500lcdt) that connected by a USB cable to my computer, reports various power supply and failure information, as shown in the panel below.
b9f0ea34-d654-4b26-b968-4f8cf0a4e0c1-image.png
afa73913-5318-42c6-af9a-02d36e2e6ea1-image.png
The application I installed runs Java and opens the console to send the information.
Has anyone tried to bring the information into the MSR? Or even a VeraPlus?
Thanks.
Hi
I am stuck trying to extract the "SignalLevel" value which is the WIFI Signal Strength off my Foscam via a http request. The request returns this, which isn't Json formatted data I don't think.
<CGI_Result> <result>0</result> <SignalLevel>52</SignalLevel> </CGI_Result>I created a rule that saves the http response into a Global Expression
872f534e-5899-43e4-ade5-f2b8ab54a412-image.png
But I am struggling to work out what is the correct expression code to extract the value?
Can anyone more knowledgeable please help?
Thank you.
Hi
Today in MSR alerts I see 100 alerts saying that the devices on my Ezlo controllers no longer exist.
They do still exist. I checked some of these devices in the Ezlo Online API Tool doing a hub.devices.list query.
I then checked that the ID numbers were the same for the devices and they are the same device ID's still.
So my devices do still exist with the same device ID's it seems.
So not sure why MSR has over 100 alerts saying otherwise.
9d7c4732-40f3-4551-b8d4-e24e0e8dda10-image.png
I then looked in Entities selected my Ezlo Plus controller and it listed all my devices OK I think and nothing looked out of place.
Thanks
I just discovered there is a secret webpage I didn't know about for my Asus Merlin router that when I run it in a browser gives me all the routers current temperatures.
http://192.168.0.1/ajax_coretmp.asp
249c6e47-38ae-4812-b149-da1fd22072cb-image.png
I have to be logged into the router to access it however.
I tried to create an MSR rule with a HTTP request to store all this data in to a local expression.
But it looks like its not passing or accepting the username and password for my router, I tried Basic and Digest.
I just get this back in the local expression in the rule.
(string) "<HTML><HEAD><script>window.top.location.href='/Main_Login.asp';</script>\n</HEAD></HTML>\n"The data returned in the browser looked like this:
curr_coreTmp_wl0_raw = "43°C"; curr_coreTmp_wl0 = (curr_coreTmp_wl0_raw.indexOf("disabled") > 0 ? 0 : curr_coreTmp_wl0_raw.replace("°C", "")); curr_coreTmp_wl1_raw = "49°C"; curr_coreTmp_wl1 = (curr_coreTmp_wl1_raw.indexOf("disabled") > 0 ? 0 : curr_coreTmp_wl1_raw.replace("°C", "")); curr_coreTmp_wl2_raw = "disabled"; curr_coreTmp_wl2 = (curr_coreTmp_wl2_raw.indexOf("disabled") > 0 ? 0 : curr_coreTmp_wl2_raw.replace("°C", "")); curr_coreTmp_wl3_raw = "disabled"; curr_coreTmp_wl3 = (curr_coreTmp_wl3_raw.indexOf("disabled") > 0 ? 0 : curr_coreTmp_wl3_raw.replace("°C", "")); curr_cpuTemp = "66.495"; fanctrl_info = "";I know how to do this for webpages and sites that give back JSON data, however this data I am not sure what format it is, Javascript ? The page is a .asp ajax page.
So not sure if this is possible ?
I really wanted to monitor the CPU temps on my router as its been getting very hot, even with extra USB fans now on the back of it. Think its a firmware bug maybe in the Asus firmware as others have been talking about it online.
Thanks.
So, I've just installed two venetian blinds and I want to control their lamellas with Fibaro FGRM222.
According to the old place, this could be accomplished with some Zwave raw messages and I'm OK with it.
Since I've moved (mostly) everything virtual in MSR, I'm now ready to add a virtual device with actions.
Looking at the docs, it seems possible, but it's not documented. @toggledbits any hints? I want to define a cover and send Zwave commands to the existing x_vera_svc_micasaverde_com_ZWaveNetwork1.SendData action. Thanks.
@toggledbits,
I'm not sure if this is something you've looked into (or something I'm completely missing in the installation documentation), but would it be possible to add user authentication to the MSR frontend? On my Home Assistant instance, I have external access set up using Cloudflare, and I can add other instances within my network to that Cloudflare configuration, but without any sort of authentication to access MSR, I obviously can't open it to the internet, unless I want random people screwing around with my home automation.
I've just updated to 23218 and now I'm getting this:
27a959bd-486b-41ae-8137-c5f456044a82-image.png
There seems to be related to global variables. The first and last ones are global reactions, while the second is a normal one. All are setting global variables.
Hey @toggledbits was wondering if we could have the latest reactor version number as an attribute for the Reactor System entity. It would be useful for me so I can use it for my wget command to automatically download the latest .tar.gz file. Ideally the complete version number like this 23171-85463e02.
@toggledbits ,
I was just taking a look at some of my scheduling to adjust for a new work schedule, and I noticed that the checkboxes for days of the week appear to be off.
75e1b936-1693-4db1-8ed5-ecfae57d3abf-image.png
if you'll notice in the screenshot, Sun - Tuesday are good, but then Sat comes next, then Wed, then Sat again, then Wed again. No TGIF in this week.
I know this isn't how I set up the rules (which would explain the behavior I was investigating).
Edit: BTW, I checked all my Week Day entries, and they are all like this.
I'm running latest-23196-40ef07f4 on Fedora 37 Server w/o Docker.
Reactor (Multi-System/Multi-Hub) Announcements
-
Build 21228 has been released. Docker images available from DockerHub as usual, and bare-metal packages here.
- Home Assistant up to version 2021.8.6 supported; the online version of the manual will now state the current supported versions;
- Fix an error in
OWMWeatherController
that could cause it to stop updating; - Unify the approach to entity filtering on all hub interface classes (controllers); this works for device entities only; it may be extended to other entities later;
- Improve error detail in messages for
EzloController
during auth phase; - Add
isRuleSet()
andisRuleEnabled()
functions to expressions extensions; - Implement
set
action forlock
andpassage
capabilities (makes them more easily scriptable in some cases); - Fix a place in the UI where 24-hour time was not being displayed.
-
T toggledbits locked this topic on
-
Build 21236
- Prototypical support for Ezlo Atom and PlugHub in EzloController;
- Be less aggressive with module loading in app.js for more permissive first-time startup;
- Simplify reaction stop for child tasks;
- Numerous small bug fixes/gnats in UI;
- Documentation updates and corrections.
Note that Generation 1 Atoms are supported in this version, but only through the remote access API (so it's cloud-dependent). Gen2 and later Atoms/PlugHubs can run with local/LAN access (no cloud dependency).
-
Build 21237
- EzloController: Tone down the warnings about disabling certain ciphers for Atom; that's only applicable to local access (don't need warnings when remote access is used).
- PR 0000245: Fixed an issue with a hang caused by the attempt to actively stop a reaction that is blocked waiting for a child reaction (a reaction group) to finish.
- PR 0000252: Enable power_switch capability for siren category on eZLO hubs.
-
Build 21239
- Expression functions
strftime()
andformat()
added. See the updated documentation. - Direct browser (via HTML attributes) to disable spellcheck on various fields where it's inappropriate (e.g. HTTP Request action URL, headers and data).
- Updated to latest lexpjs, which has itself been updated to (hopefully) support Unicode alphabetic characters in identifier names.
- Allow expressions to have Unicode characters in their names (parity with lexpjs).
- Additional preparation for localization in UI and engine. For example, non-US users should notice that lists sorted alphabetically now appear in a more natural order for their configured language, rather than "ASCII" (character code ordinal) sorting.
- Expression functions
-
Build 21243
- HubitatController: Resolve exception thrown when attempting to set mode.
- HassController: Blessed Hass version 2021.8.8
-
Build 21256
- Fix issue with HTTP Request action post body causing exception;
- The maximum wait time for controllers to become available when Reactor starts is now configurable via the
startup_wait
key in theengine
section ofreactor.yaml
. Users of the similarly-namedstartup_delay
, which is a fixed delay, are encouraged to usestartup_wait
instead. - The Rule detail display in the Rule Sets list was not correctly showing some non-primitive values (objects and arrays) correctly.
- HassController: Blessed Hass version 2021.9.6
-
Build 21258
- Warning and clear when changing operator on condition that would clear condition options.
- Fix Vera importer geotag handling.
- VeraController: add binary
x_vera_geotag.in_region
attribute. - HassController: implement
x_hass_system.call_service
on system entities.
-
Build 21267
- Change the evaluation algorithm for global variables to make them reactive. That is, global variables are now automatically reevaluated when dependent global variables or entities (from use
getEntity()
) change. They will also drive evaluation of rules when a rule-based expression or trigger condition references a global variable that changes. - A new version of lexpjs supporting the above changes is being used; this version includes local scope for the interior of most statements, which may break some expressions (in particular, those that breach etiquette and create variables inside loops but access them outside).
- First version with localization capability. Users interested in localizing the user interface for their language/country can read the how-to documentation.
Expressions Changes
This release comes with some significant changes to the operation of global expressions in particular, and some expression tweaks throughout.
First, the expression language has been given the concept of scope; specifically, the statements
each
,first
,do...done
,if then else endif
now have local scopes for their interior expressions. Variables created in the interior of these statements are not visible outside of these statements (they are implicitlylocal
), unless the variable already exists in an enclosing scope. This is probably not going to cause anyone much heartburn on existing expressions with these statements. I don't see many of you pushing the limits in this area of the expression language that would make it an issue. We'll handle exceptions as they come up.Second, and more importantly, Reactor now evaluates and treats global variables differently, and I think you'll agree this is a good change and brings more useful, even expected, behavior. The details:
- Global variables are now reactive; for example, when a global variable that contains a
getEntity()
expression detects that the entity is modified, the global variable's expression is automatically recalculated. In previous versions of MSR, the variable was not re-evaluated until a rule requested the global variable's value. - Global variables are recalculated when they reference another global variable, and that global variable changes (for whatever reason). If, for example, you have a rule that periodically fetches JSON data from a web site and stores the response in a global variable, and other global variables that draw on that data, those dependent variables will be reevaluated when the variable holding the response is modified.
- Rules will be reactive to global variable changes that do not involve entities. Previously, MSR rules would only re-evaluate on a change to a global variable if that variable involved a
getEntity()
expression. In fact, Reactor was not reacting to the variable change, it was reacting to the entity change. Now, Reactor truly reacts to the global variable change, which means it will also respond to the change in values of case #2 above -- if a rule is dependent on a global variable, any change to that global variable for any reason will cause the rule to re-evaluate, and based on its conditions, possibly change state. - You can disable the dependency and auto-recalculation of a global variable, because sometimes its useful to do that.
Given these changes, it is pretty easy to create reference loops. There's nothing stopping you from creating a variable A and compute its value from B, and then create B and have it set its value based on A. This will get you loop errors and a stop on evaluation; it won't be subtle. Similar abuses between global variables and rules and the overzealous use of Set Variable actions can also lead to rule loops throttling. Any time you use Set Variable in a rule-based reaction where that rule also uses the variable in triggers, you are running the risk of such a loop, so be careful.
Reactor for Vera doesn't have global variables, so at the time it was implemented as a new feature of MSR, simplifying assumptions were made. I hope you'll agree this is a useful evolution of their behavior.
Docs are updated, so make sure you check them out.
Localization
This is the first release that has the bulk of my recent work on localization. I'm barely passable at German, certainly not enough so to do a translation, and know enough Spanish to order a beer and get slapped by members of the opposite sex, but that's about it for me for languages (I used to teach Latin in an elementary school, but I doubt there's much demand for that translation). So while the system now recognizes the language configuration of your browser and host system automatically, it won't be able to present in any language other than US English because that's the only string file available. If any native speakers of other languages (including UK and Australian English) would like to step up and do a localization, let me know — that would be your work product to own and distribute as an add-on (e.g. through Github, etc.) to the community, and optionally include in Reactor distributions if you're willing (as a redistributed/third-party work with attribution on the About page). The US English file, which is the reference for all strings and starting point for translations, can be found in this Github repository: https://github.com/toggledbits/msr-localization
- Change the evaluation algorithm for global variables to make them reactive. That is, global variables are now automatically reevaluated when dependent global variables or entities (from use
-
Build 21270
- Fix an issue with EzloController not processing device match expressions correctly on newer lexpjs.
- Add
matchEntities()
function (see docs) - Further localization improvements and fixes.
-
Build 21275
NOTICE: This revision incorporates a breaking change to the recently-introduced
matchEntities()
expression function. See below.- VeraController: Devices of type
...:FloodSensor1
now use theleak_sensor
(specific) capability; they maintain their previous use of the genericbinary_sensor
capability for backward compatiblity with existing automations. - EzloController: Fix a stack overflow that could occur under certain boundary conditions.
- OWMController: Fix a broken context initialization for
lexpjs
causing user-configured "extra data" fields with expressions to throw a soft error and report an incorrect (null) value. - Improve interaction between Engine and UI when global expressions are created: they are now given an initial evaluation and their dependency tree built at that time. Previously, a rule reference or manual UI interaction (e.g. "Try Now" button press) was required to initialize it.
- Improve interaction between Engine and UI when global variables are deleted: the deleted variable is now removed from the Engine's evaluation dependency tree to avoid spurious "Expression Not Found" exceptions in the log for the then-deleted variable. These would stop after a restart, because the dependency tree was rebuilt at that time, but it should not take a restart to fix the tree.
- Fix an error in the parameters of the
isRuleSet
andisRuleEnabled
expression functions that would cause a constructor error when global expressions using these functions were initialized at system startup, before the Engine had loaded the rules. This error was exposed by the changes made to support global variable dependency graphs. - When deleting system objects, wait until all notifications have been given and returned, to assure that the object being deleted is not actually deleted before the notified object has a chance to do whatever it needs to do. The former behavior had no adverse effect; this was a change to gain new behavior necessary to support the global variable dependency tree update on delete mentioned above.
- Fix a bug in the recording of the timestamp of modification of a global variable; it was using real time rather than virtual time, so when the engine was configured for "Test Time", it was recording incorrect (i.e. non-test time) values.
- The check for circular dependencies in the evaluation of global variables was relaxed to allow self-references (e.g.
series = push( series, value )
). - BREAKING: The
matchEntities()
function now returns only a list of canonical IDs for matched entities. You will need to use thegetEntity()
function to fetch current data for the entity if you need it. This was done because the previously-returned list of entity objects could grow "stale" (the data in the objects reflects the state at the time, not ever-changing and ongoing state as attributes change). And yes, this is exactly the kind of thing I said I try to avoid, but sometimes you just don't know all the side-effects until you see the thing in the wild. Sorry, everyone. - New expression functions:
fileExists()
,fileRead()
,fileWrite()
. See the docs. - Incorporates a hotfix made to 21270 for the
SetVariable
action on Vera hubs.
- VeraController: Devices of type
-
Build 21277
- HubitatController: Fix a problem with setting RGB color.
- Fix an issue causing "Object not found" exceptions when evaluating a
getEntity()
call in a subcontext of a rule-based expression (e.g. within aneach
loop). - Fix a number of issues required for localization (thank you, @Crille ).
-
T toggledbits pinned this topic on
-
Build 21281
- In prior versions, a reference to a variable (rule or global) that does not exist would cause an alert and "fail soft" by returning null. This (old) behavior actually hides such errors and makes them more difficult to locate, and generally speaking, you shouldn't be referring to things that don't exist. The new behavior is to throw an exception during the evaluation, which will more clearly highlight the error and make it easier to find and correct. If for some reason you still need the old behavior, you can set
deprecated_undefined_global_soft_fail
to true in theengine
section ofreactor.yaml
, and such errors will soft-fail as before. The old behavior is deprecated, however, and will be removed in a few months. - PR 0000260: Fix failure of HTTP Request (really underlying variable handling) to notice change in array of objects returned by request.
- PR 0000259: Need to more clearly highlight variables with errors in the expression editor. Failing expressions now receive orange highlight bars.
- Add
typeof()
expression function. - Additional localization changes.
- New HTTP API additions: add alert, dismiss alert
- Bless Home Assistant to 2021.10.0
- In prior versions, a reference to a variable (rule or global) that does not exist would cause an alert and "fail soft" by returning null. This (old) behavior actually hides such errors and makes them more difficult to locate, and generally speaking, you shouldn't be referring to things that don't exist. The new behavior is to throw an exception during the evaluation, which will more clearly highlight the error and make it easier to find and correct. If for some reason you still need the old behavior, you can set
-
Build 21286
- InfluxFeed now supports InfluxDB version 1.8+ and 2.0. InfluxDB versions prior to 1.8 are no longer supported. Please see the documentation for configuration details, as the configuration requirements for 1.x and 2.0 are different. Users of InfluxFeed also need to
npm update --nosave
if they are bare-metal installed. - The behavior of rules when throttled has been improved. Note, however, that any time a rule is throttled, events could be missed (e.g. if a light that is a dependency of a rule is turned on and back off quickly when throttled, the rule may not "see" the on event).
- Ongoing documentation updates.
- Bless Hass to 2021.10.4
- First version to contain a foreign language translation, Swedish! Thank you @Crille for your work on the translation and helping me get localization working.
- Address possible evaluation loop while using SetVariable in a rule reaction on a rule variable, where the value is an expression that contains a
getEntity()
reference. - Documentation updates
- Further i18n fixes (thank you @Crille)
- InfluxFeed now supports InfluxDB version 1.8+ and 2.0. InfluxDB versions prior to 1.8 are no longer supported. Please see the documentation for configuration details, as the configuration requirements for 1.x and 2.0 are different. Users of InfluxFeed also need to
-
Build 21292
- PR 0000262: Fix an error in handling date/time conditions with M/D across new year boundary.
- PR 0000261: HubitatController: Resolve an order of operations issue preventing extended attributes from being assigned as primary in local config.
- PR 0000258: (i18n) Fix string reference for condition option
- HassController: treat controller reporting NOT_RUNNING as an error and disconnect/retry
- Bless Hass to 2021.10.6
-
Build 21294
- PR 000263: Fix spurious log messages after deleting a global expression that contained a
getEntity()
. - Fixed a UI issue with an expression being sorted to the top of the list then ending up at the bottom.
- The expression function
strftime()
is now locale-aware. - Docker containers are now running on nodejs version 16.
- PR 000263: Fix spurious log messages after deleting a global expression that contained a
-
Build 21297
- PR 0000264: Fix an error in the rule editor on a particular value test used in several places that could cause the entered value of a condition option to not be saved (most notable on the "repeats within" field, but also found in two other locations).
- Update to lexpjs 21296 to get
sort()
function for arrays, and user-definable functions from expressions. Default sort in Reactor Engine for the implemention of expressionsort()
function is (host) locale-aware. Please refer to the documentation for details.
-
Build 21306
Note: this release includes changes to package dependencies. Bare-metal users please run
npm update --no-save
to resolve.- HTTP Request action now has explicit control over HTTP auth, with type selection (none, Basic, Digest) and credential fields (username and password).
- PR 0000265: Correct a string that is not translated in the alerts status panel.
- i18n: Found an additional string that was not localizable.
- The HTTP Request action will now suppress all alerts for any error it encounters, not just HTTP errors, if the "no error alerts" option is checked.
- Update lexpjs to get new array set functions:
arrayIntersection()
,arrayUnion()
,arrayDifference()
,arrayExclusive()
andarrayConcat()
.
-
Build 21307
- PR 0000266: An issue introduced at 21306 saves certain substitutions with the wrong data type.
- HubitatController: Resolve (hopefully) a race condition at startup.
-
Build 21313
- Bless Home Assistant to version 2021.11.2
- New system capability
button
to express state of a single button or contact closure. - PR 0000268: Sun condition shows "not between" as "between" in rule detail card.
- Add
/alert/:tag
HTTP API endpoint (DELETE method); this deprecates/alert/:tag/dismiss
.
-
Build 21321
- HassController: Additional updates for event handling (
init
support). - HubitatController: Map additional capabilities.
- Internal support additionals for MQTTController
MQTTController - Build 21321
- Fix incorrect scope reference is tasmota_generic_relay payload.
- Additional helpful debug.
Upgrading MQTTController, which is a standalone/add-on component, is the same procedure as initial installation.
- HassController: Additional updates for event handling (