I’m cleaning up my HA instance and most everything is working well except for the Node-RED app v22.0.1; Node-RED v5.0.2. I started having issues since HA 2026.8.x (1 or 2). Upon an HA system restart, everything comes back up just fine.
Here’s a successful startup log. I see some issues with some entities and nodes that I don’t entirely understand. input_boolean.pool_open is an existing entity with a valid state upon startup, so I’m not sure why that’s an issue… could not be found in cache? Also the tablet battery level entity…?
There are also some logs related to all of my cronplus nodes. Not sure why those are an issue suddenly.
16 Aug 08:16:55 - [info]
Welcome to Node-RED
===================
16 Aug 08:16:55 - [info] Node-RED version: v5.0.2
16 Aug 08:16:55 - [info] Node.js version: v24.18.1
16 Aug 08:16:55 - [info] Linux 6.18.39-haos x64 LE
16 Aug 08:16:56 - [info] Loading palette nodes
16 Aug 08:16:57 - [info] Node-RED Contrib Theme Collection version: v5.0.1
16 Aug 08:17:06 - [info] node-red-contrib-home-assistant-websocket v0.80.3 nodes initialized
16 Aug 08:17:06 - [info] Settings file : /etc/node-red/config.js
16 Aug 08:17:06 - [info] Context store : 'default' [module=memory]
16 Aug 08:17:06 - [info] User directory : /config/
16 Aug 08:17:06 - [warn] Projects disabled : editorTheme.projects.enabled=false
16 Aug 08:17:06 - [info] Flows file : /config/flows.json
16 Aug 08:17:06 - [info] Server now running at http://127.0.0.1:46836/
16 Aug 08:17:07 - [info] Starting flows
16 Aug 08:17:08 - [info] Started flows
16 Aug 08:17:12 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:17:17 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:17:18 - [error] [api-current-state:Pool open?] InputError: Entity could not be found in cache for entityId: input_boolean.pool_open
16 Aug 08:17:19 - [error] [api-current-state:Pool open?] InputError: Entity could not be found in cache for entityId: input_boolean.pool_open
16 Aug 08:17:19 - [error] [api-current-state:Pool open?] InputError: Entity could not be found in cache for entityId: input_boolean.pool_open
16 Aug 08:17:19 - [error] [api-current-state:Pool open?] InputError: Entity could not be found in cache for entityId: input_boolean.pool_open
16 Aug 08:17:20 - [error] [api-current-state:First Floor Tablet Battery <= 20%?] InputError: Entity could not be found in cache for entityId: sensor.first_floor_tablet_battery
16 Aug 08:17:20 - [error] [api-current-state:First Floor Tablet Battery >= 80%?] InputError: Entity could not be found in cache for entityId: sensor.first_floor_tablet_battery
16 Aug 08:17:20 - [error] [api-current-state:First Floor Tablet Battery <= 20%?] InputError: Entity could not be found in cache for entityId: sensor.first_floor_tablet_battery
16 Aug 08:17:20 - [error] [api-current-state:First Floor Tablet Battery >= 80%?] InputError: Entity could not be found in cache for entityId: sensor.first_floor_tablet_battery
16 Aug 08:17:20 - [error] [api-call-service:Switch Tablet to Default View] NoConnectionError
16 Aug 08:17:20 - [error] [api-call-service:Switch Tablet to Default View] NoConnectionError
16 Aug 08:17:20 - [error] [api-call-service:Reset] NoConnectionError
16 Aug 08:17:20 - [error] [ha-sensor:Interior Lights Time Remaining] NoConnectionError
16 Aug 08:17:29 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:17:34 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:17:38 - [error] [api-current-state:xxxx at home?] InputError: Entity could not be found in cache for entityId: sensor.status_xxxx
16 Aug 08:17:38 - [error] [api-current-state:xxxx at home?] InputError: Entity could not be found in cache for entityId: sensor.status_xxxx
16 Aug 08:17:39 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:17:44 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:17:49 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:17:54 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:17:59 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:18:04 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:18:08 - [error] [api-current-state:Pool open?] InputError: Entity could not be found in cache for entityId: input_boolean.pool_open
16 Aug 08:18:09 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:18:14 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:18:19 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:18:24 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:18:29 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:18:34 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:18:39 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:18:44 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:18:49 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:18:54 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:18:59 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:19:04 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:19:08 - [error] [api-current-state:Pool open?] InputError: Entity could not be found in cache for entityId: input_boolean.pool_open
16 Aug 08:19:08 - [error] [api-call-service:Run Watchman Report] NoConnectionError
16 Aug 08:19:09 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:19:17 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:19:22 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:19:27 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:19:32 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:19:37 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:19:42 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:19:47 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:19:57 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:20:03 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:20:08 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:20:09 - [error] [api-current-state:Pool open?] InputError: Entity could not be found in cache for entityId: input_boolean.pool_open
16 Aug 08:20:14 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:20:23 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:20:28 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:20:34 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:20:39 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:20:44 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:20:49 - [info] [server:Home Assistant] Connecting to http://supervisor/core
16 Aug 08:20:52 - [info] [server:Home Assistant] Connected to http://supervisor/core
16 Aug 08:21:14 - [info] [cronplus:Weekday & Weekend PM] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
16 Aug 08:21:14 - [info] [cronplus:Weekday & Weekend AM] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
16 Aug 08:21:15 - [info] [cronplus:Daily at 7am] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
16 Aug 08:21:15 - [info] [cronplus:Daily May - Nov at 7am] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
16 Aug 08:21:15 - [info] [cronplus:Daily at 9am & 10pm] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
16 Aug 08:21:27 - [info] [cronplus:Weekday & Weekend PM] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
16 Aug 08:21:28 - [info] [cronplus:Weekday & Weekend AM] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
16 Aug 08:21:28 - [info] [cronplus:Daily at 7am] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
16 Aug 08:21:28 - [info] [cronplus:Daily May - Nov at 7am] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
16 Aug 08:21:28 - [info] [cronplus:Daily at 9am & 10pm] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
16 Aug 08:21:28 - [info] [ping:Ping] ping - Host 'Home' process timeout - sending SIGINT
Node-RED crashes after some time… between an hour and several hours. Here’s a post-crash log, when the app is down:
16 Aug 08:07:09 - [info] Node-RED version: v5.0.2
16 Aug 08:07:09 - [info] Node.js version: v24.18.1
16 Aug 08:07:09 - [info] Linux 6.18.39-haos x64 LE
16 Aug 08:07:09 - [info] Loading palette nodes
2026/08/16 08:07:11 [error] 512#512: *1 connect() failed (111: Connection refused) while connecting to upstream, client: 172.30.32.2, server: a0d7b954-nodered, request: "GET / HTTP/1.1", upstream: "http://127.0.0.1:46836/", host: "10.0.0.57:8123", referrer: "http://10.0.0.57:8123/a0d7b954_nodered"
16 Aug 08:07:13 - [info] Node-RED Contrib Theme Collection version: v5.0.1
16 Aug 08:07:23 - [info] node-red-contrib-home-assistant-websocket v0.80.3 nodes initialized
16 Aug 08:07:24 - [info] Settings file : /etc/node-red/config.js
16 Aug 08:07:24 - [info] Context store : 'default' [module=memory]
16 Aug 08:07:24 - [info] User directory : /config/
16 Aug 08:07:24 - [warn] Projects disabled : editorTheme.projects.enabled=false
16 Aug 08:07:24 - [info] Flows file : /config/flows.json
16 Aug 08:07:25 - [info] Server now running at http://127.0.0.1:46836/
16 Aug 08:07:28 - [info] Starting flows
16 Aug 08:07:35 - [info] Started flows
16 Aug 08:07:35 - [info] [server:Home Assistant] Connecting to http://supervisor/core
[08:07:39] INFO: Service Node-RED exited with code 256 (by signal 9)
There are no available palette updates. Any help with deciphering these problems would be helpful!
EDIT: I saw this thread re: cronplus nodes, but my browser-mod devices have a reasonable number of entities (~13 each).
EDIT 2: Checked this thread re: entities & caching, but I have no init_commands in my NR app.
EDIT 3: the System Time Change Detected log entries persist and repeat. I suspect that’s my main issue.
How do all of those entities look in Settings - Devices - Entities and Settings and Devices - Helpers? Are they defined in configuration.yaml or using the Helpers GUI? Make sure they all have the same entity_id’s that Node Red is referencing. I had some issues where a few Helper entities were deleted when I upgraded to 2026.8, and several were automatically renamed to include the new (forced) area entity ID format. Node Red threw quite a few errors with the missing/renamed entities.
Check your system time and make sure it’s correct. Not sure when it happened, but I’m now noticing multiple entries under Settings - System - Home Information - Region for the same time zone. I’m assuming there’s a reason for 18 different entries for -5:00 Eastern Time, but I have no idea why that’s needed. I continue to get a log entry for a time mismatch between Home Assistant and a POE camera, but both show the exact, correct times.
Make sure you only have one homeassistant server listed in Node Red - Configuration Nodes.
Thanks for looking. Good questions… first thing I checked. Most of my template sensors are defined in YAML… have been for several years. One day, I’ll move things to the UI. But, for now…
sensor.first_floor_tablet_battery is coming from Fully Kiosk. Same entity_id:
input_boolean.pool_open I set up as a helper through the UI. Same entity_id. Current value is on. I toggle this one manually twice a year.
sensor.status_xxxx is defined in YAML. It resolves zone name, locality, and some other stuff to give me Lovelace-friendly location text for each user (xxxx, yyyy, etc.). They have identical state definitions:
{% if is_state('input_select.xxxx', 'Just Left')
or is_state('input_select.xxxx', 'Just Arrived')
or is_state('input_select.xxxx', 'Home') %}
{{ states('input_select.xxxx') }}
{% elif is_state('person.xxxx','Away') or is_state('person.xxxx','not_home') %}
{{ state_attr('sensor.xxxxs_iphone_geocoded_location', 'Locality') }}
{% else %}
{{ states('person.xxxx') }}
{% endif %}
I don’t know why this one would be called out in the log and not the other two. They all have string values. This one is currently “Home” and has been all day, through restarts.
So… not sure why they would bring Node-RED any problems!
this is a bit tricky I could write down the complete way if you really fancy.
In short: backup up your flows, delete them delete the global setting of the contrib, remove the contrib, restart node-red and reimport the flows.
You are then getting your flows with greyed out node which would require that contrib.
In case node-red runs … you know about where to look for the problem.
Most easy would be to stop node-red, backup the complete /addon_configs/a0d7b954_nodered/ folder, then delete flows, settings nad remove the contrib.
And if all worked on after testing stop node-red and replace everything you backed up by overwriting things. This would not only establisj the contrib but also all it’s settings.
But I’m more but 100% sure that @mightybosstone isn’t too wrong.
If you try to reconfigure all of you nodes belonging to that contrib I bet you’ll stumble across one or more where you need to use a differnt entity_id or name but you use right now.
It looks like every action node is affected. I’m inclined to think the websocket is to blame. I’ve backed up my flows… may need to take a clean slate approach here.
Edit any action or current state node and click on the pencil icon next to the homeassistance server drop-down. Make sure the “Using the Home Assistant Add-on” option is checked (assuming you’re using the Node Red Add-on app).
I am! I did a quick survey and opened 6-7 action nodes and they all have that box checked.
I’ve spent the past few hours cleaning up my templates.yaml, adding availability to a bunch of sensors.
After a restart, all looks normal in the UI. Logs look better; there are no more entries for entities not found in cache. There are still a few popping up for system time change.
Things are looking up. Not perfect, but functional, at least. I’ll know more tomorrow; every night for the last week or so, NR has failed overnight and needed a kick in the morning to get working again.
I woke up this morning and… NR was down again! 502 Bad Gateway. Even after all the cleanup yesterday. Here’s the log:
Welcome to Node-RED
===================
17 Aug 07:32:14 - [info] Node-RED version: v5.0.2
17 Aug 07:32:14 - [info] Node.js version: v24.18.1
17 Aug 07:32:14 - [info] Linux 6.18.39-haos x64 LE
17 Aug 07:32:14 - [info] Loading palette nodes
17 Aug 07:32:15 - [info] Node-RED Contrib Theme Collection version: v5.0.1
17 Aug 07:32:23 - [info] node-red-contrib-home-assistant-websocket v0.80.3 nodes initialized
17 Aug 07:32:23 - [info] Settings file : /etc/node-red/config.js
17 Aug 07:32:23 - [info] Context store : 'default' [module=memory]
17 Aug 07:32:23 - [info] User directory : /config/
17 Aug 07:32:23 - [warn] Projects disabled : editorTheme.projects.enabled=false
17 Aug 07:32:23 - [info] Flows file : /config/flows.json
17 Aug 07:32:23 - [info] Server now running at http://127.0.0.1:46836/
17 Aug 07:32:23 - [info] Starting flows
17 Aug 07:32:25 - [info] Started flows
17 Aug 07:32:28 - [info] [server:Home Assistant] Connecting to http://supervisor/core
17 Aug 07:32:29 - [info] [server:Home Assistant] Connected to http://supervisor/core
[07:32:37] INFO: Service Node-RED exited with code 256 (by signal 9)
[07:32:38] INFO: Starting Node-RED...
> start
> node node_modules/node-red/red.js --settings /etc/node-red/config.js
17 Aug 07:32:40 - [info]
Welcome to Node-RED
===================
17 Aug 07:32:40 - [info] Node-RED version: v5.0.2
17 Aug 07:32:40 - [info] Node.js version: v24.18.1
17 Aug 07:32:40 - [info] Linux 6.18.39-haos x64 LE
17 Aug 07:32:41 - [info] Loading palette nodes
s6-rc: info: service legacy-services: stopping
s6-rc: info: service nginx: stopping
s6-rc: info: service legacy-services successfully stopped
[07:32:44] INFO: Service NGINX exited with code 0 (by signal 0)
s6-rc: info: service init-nginx: stopping
s6-rc: info: service nodered: stopping
s6-rc: info: service nginx successfully stopped
s6-rc: info: service init-nginx successfully stopped
[07:32:44] INFO: Service Node-RED exited with code 0 (by signal 0)
s6-rc: info: service init-nodered: stopping
s6-rc: info: service nodered successfully stopped
s6-rc: info: service init-customizations: stopping
s6-rc: info: service init-nodered successfully stopped
s6-rc: info: service legacy-cont-init: stopping
s6-rc: info: service init-customizations successfully stopped
s6-rc: info: service fix-attrs: stopping
s6-rc: info: service base-addon-log-level: stopping
s6-rc: info: service legacy-cont-init successfully stopped
s6-rc: info: service fix-attrs successfully stopped
s6-rc: info: service base-addon-banner: stopping
s6-rc: info: service base-addon-log-level successfully stopped
s6-rc: info: service s6rc-oneshot-runner is marked as essential, not stopping it
s6-rc: info: service base-addon-banner successfully stopped
/package/admin/s6-overlay/libexec/preinit: info: container permissions: uid=0 (root), euid=0, gid=0 (root), egid=0
/package/admin/s6-overlay/libexec/preinit: info: /run permissions: uid=0 (root), gid=0 (root), perms=oxorgxgruxuwur
/run/s6/basedir/scripts/rc.init: warning: defining user bundles in /etc/s6-overlay/s6-rc.d is deprecated, please define them in /etc/s6-overlay/user-bundles.d instead
s6-rc: info: service s6rc-oneshot-runner: starting
s6-rc: info: service base-addon-banner: starting
s6-rc: info: service s6rc-oneshot-runner successfully started
-----------------------------------------------------------
App: Node-RED
Flow-based programming for the Internet of Things
-----------------------------------------------------------
App version: 22.0.1
You are running the latest version of this app.
System: Home Assistant OS 18.2 (amd64 / qemux86-64)
Home Assistant Core: 2026.8.2
Home Assistant Supervisor: 2026.07.5
-----------------------------------------------------------
Please, share the above information when looking for help
or support in, e.g., GitHub, forums or the Discord chat.
-----------------------------------------------------------
s6-rc: info: service fix-attrs: starting
s6-rc: info: service base-addon-log-level: starting
s6-rc: info: service base-addon-banner successfully started
s6-rc: info: service fix-attrs successfully started
Log level is set to INFO
s6-rc: info: service legacy-cont-init: starting
s6-rc: info: service base-addon-log-level successfully started
s6-rc: info: service init-nginx: starting
s6-rc: info: service init-customizations: starting
s6-rc: info: service legacy-cont-init successfully started
s6-rc: info: service init-nodered: starting
s6-rc: info: service init-customizations successfully started
s6-rc: info: service init-nginx successfully started
@justone suggested moving my flows to MQTT. I will probably try that, but I’m not sure that’ll solve my problem.
For the NR app, I have the watchdog set so it’ll restart the app if it crashes. However… the UI still isn’t available and none of my flows run when it’s in this state. I can go over to the info page, stop the app, and “Error” appears in the header. I can then attempt to start it, but won’t successfully start:
[07:40:36] INFO: Starting Node-RED...
> start
> node node_modules/node-red/red.js --settings /etc/node-red/config.js
17 Aug 07:40:38 - [info]
Welcome to Node-RED
===================
17 Aug 07:40:38 - [info] Node-RED version: v5.0.2
17 Aug 07:40:38 - [info] Node.js version: v24.18.1
17 Aug 07:40:38 - [info] Linux 6.18.39-haos x64 LE
17 Aug 07:40:38 - [info] Loading palette nodes
17 Aug 07:40:40 - [info] Node-RED Contrib Theme Collection version: v5.0.1
17 Aug 07:40:47 - [info] node-red-contrib-home-assistant-websocket v0.80.3 nodes initialized
17 Aug 07:40:47 - [info] Settings file : /etc/node-red/config.js
17 Aug 07:40:47 - [info] Context store : 'default' [module=memory]
17 Aug 07:40:47 - [info] User directory : /config/
17 Aug 07:40:47 - [warn] Projects disabled : editorTheme.projects.enabled=false
17 Aug 07:40:47 - [info] Flows file : /config/flows.json
17 Aug 07:40:47 - [info] Server now running at http://127.0.0.1:46836/
[07:40:48] INFO: Starting NGinx...
17 Aug 07:40:48 - [info] Starting flows
17 Aug 07:40:50 - [info] Started flows
17 Aug 07:40:53 - [info] [server:Home Assistant] Connecting to http://supervisor/core
17 Aug 07:40:53 - [info] [server:Home Assistant] Connected to http://supervisor/core
17 Aug 07:41:04 - [info] [cronplus:Weekday & Weekend PM] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
17 Aug 07:41:04 - [info] [cronplus:Weekday & Weekend AM] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
17 Aug 07:41:04 - [info] [cronplus:Daily at 7am] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
17 Aug 07:41:04 - [info] [cronplus:Daily May - Nov at 7am] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
17 Aug 07:41:04 - [info] [cronplus:Daily at 9am & 10pm] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
[07:41:12] INFO: Service Node-RED exited with code 256 (by signal 9)
[07:41:13] INFO: Starting Node-RED...
> start
> node node_modules/node-red/red.js --settings /etc/node-red/config.js
17 Aug 07:41:15 - [info]
Usually an indication that Node-RED has run out of memory.
Node-RED really only crashes if it runs out of memory, and it really only runs out of memory if there is a memory leak somewhere.
Running out of memory in a node-js process is a catastrophic failure, with the node-js garbage collection thrashing and taking up most of the processor time. This would explain the cronplus node time-issues and the failure of WebSockets too, as with garbage collection dominant no other processes have processor resource to run. Eventually there is not enough memory to run the garbage collection, and with no garbage collection, it all stops.
There are two key reasons for an NR memory leak.
you have a flow, usually a loop, that consumes resources on each loop but does not free them up [poor coding]
you have a node that consumes resources, but is not releasing them [badly behaved node]
You can check for memory issues by monitoring the RSS memory consumption, and there are various things to look out for in your flow code, but I wonder if you have problems from browser-mod…
Edit:
Of course the first question to ask you is what are you running HA on (with Node-RED as App) and does it have enough memory? Every update of HA core takes just a bit more memory, and if you are running lots of Apps and integrations, you may just be out of memory generally…
If it is a memory issue, perhaps running Node-RED with jemalloc could be helpful. If enabling it alone isn’t enough, there are several tuning options that might be worth trying.
jemalloc is available as a configurable option of the Node-RED Plus app:
I’ve been out of town all week and unable to pay any attention to this; to combat the NR issues and keep my automations running, I’ve been restarting HA a couple of times a day… definitely not sustainable.
I don’t run a VM myself so this is a bit of a guess, but if your screenshot suggests that you have a Synology NAS with 8GB of physical memory, and you have allocated 2 GB to a VM to run Home Assistant, then I assume that you are running HA on 2GB.
As I understand the current position, Home Assistant, with lots of Apps, can struggle on 2GB.
I think you are running out of memory. Node-RED will stop if it runs out of memory (and I am sure that HA will struggle too…).
Yes, that’s correct. (8GB/2GB)… what’s a more reasonable allocation? 4GB? This is the only VM I run on this NAS. Its only purpose is network storage and HA.
Doing a little digging, and it appears I have a little over 3GB of RAM headroom with what I have now. I’ll assign 3GB to HA and see if that helps.
I didn’t consider that I could be running low on memory… but it would be a logical explanation.
I can only go by the comments made elsewhere regarding issues around running a heavy HA system on 2GB.
The HA Green I believe has 4GB. Yes you can, apparently, run HA on 2GB, but not with a lot of Apps and heavy processing (such as video work).
Every update of HA is just a little bit bigger… Around 1.5GB for HA now I believe. NR runs in quite a small space, but does need to be able to call for extra memory from the ‘system’.
Edit:
If Sir_Goodenough can’t run HA on a VM with 2GB, then I suspect you can’t either…
Great data point. I’ve just brought everything up on 3 cores/4GB (maybe a little over-allocated on the CPU) and it’s working well.
Now… I’m still seeing some time change entries in the log:
22 Aug 11:30:06 - [info] [server:Home Assistant] Connecting to http://supervisor/core
22 Aug 11:30:06 - [info] [server:Home Assistant] Connected to http://supervisor/core
22 Aug 11:30:17 - [info] [cronplus:Weekday & Weekend PM] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
22 Aug 11:30:17 - [info] [cronplus:Weekday & Weekend AM] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
22 Aug 11:30:17 - [info] [cronplus:Daily at 7am] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
22 Aug 11:30:17 - [info] [cronplus:Daily May - Nov at 7am] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
22 Aug 11:30:17 - [info] [cronplus:Daily at 9am & 10pm] System Time Change Detected - refreshing schedules! If the system time was not changed then this typically occurs due to blocking code elsewhere in your application
In my attempts to figure this all out, I installed the Node-RED Plus add-on and found that I couldn’t add the Time Range Switch to my palette. This is the one that is appearing in the logs… but not all of that node type appear there (I have several others that don’t show above).
So, still looking for answers/fixes, I suppose. I can tell my flows are more responsive than they have been in awhile, but time will tell if NR stays up on its own.
I believe the recommendation for VM is for 2 cores at least. Node-RED is single thread, so if you have a blocking node then this will take one core, and if something else in HA is blocking there, then that takes another core.
The cronplus node will automatically reset where there is a change or drift in the system time. If this happens regularly, then I would suspect issues with the Synology VM machine system time stability. Raspberry PIs have an NTP server sync, as well as a real-time clock and (optional) battery. A VM ‘machine’ guest operating system needs to have system time updated from the host machine or NTP.
If it is a minor issue, then the cronplus node has an optional setting to modify the drift-value trigger point.
