Hey, I just installed the nodered container (wanted to try and mess with it in home assistant) but I'm having some trouble installing any nodes.  Anyone have any thoughts, here's the install log:
 

--edit - I found an issue on docker hub, the issue was hard links need to be enabled under Global Share Settings:
https://github.com/node-red/node-red-docker/issues/332

 

0 verbose cli /usr/local/bin/node /usr/local/bin/npm
1 info using [email protected]
2 info using [email protected]
3 timing npm:load:whichnode Completed in 0ms
4 timing config:load:defaults Completed in 1ms
5 timing config:load:file:/usr/local/lib/node_modules/npm/npmrc Completed in 0ms
6 timing config:load:builtin Completed in 1ms
7 warn config production Use `--omit=dev` instead.
8 timing config:load:cli Completed in 1ms
9 timing config:load:env Completed in 0ms
10 timing config:load:file:/data/.npmrc Completed in 0ms
11 timing config:load:project Completed in 1ms
12 timing config:load:file:/root/.npmrc Completed in 0ms
13 timing config:load:user Completed in 0ms
14 timing config:load:file:/usr/local/etc/npmrc Completed in 1ms
15 timing config:load:global Completed in 1ms
16 timing config:load:validate Completed in 0ms
17 timing config:load:credentials Completed in 1ms
18 timing config:load:setEnvs Completed in 1ms
19 timing config:load Completed in 9ms
20 timing npm:load:configload Completed in 9ms
21 timing npm:load:mkdirpcache Completed in 3ms
22 timing npm:load:mkdirplogs Completed in 1ms
23 verbose title npm install [email protected]
24 verbose argv "install" "--no-audit" "--no-update-notifier" "--no-fund" "--save" "--save-prefix" "~" "--production" "--engine-strict" "[email protected]"
25 timing npm:load:setTitle Completed in 1ms
26 timing config:load:flatten Completed in 3ms
27 timing npm:load:display Completed in 5ms
28 verbose logfile logs-max:10 dir:/data/.npm/_logs
29 verbose logfile /data/.npm/_logs/2023-12-03T18_08_57_891Z-debug-0.log
30 timing npm:load:logFile Completed in 6ms
31 timing npm:load:timers Completed in 0ms
32 timing npm:load:configScope Completed in 0ms
33 timing npm:load Completed in 27ms
34 timing arborist:ctor Completed in 0ms
35 silly logfile done cleaning log files
36 timing arborist:ctor Completed in 0ms
37 timing idealTree:init Completed in 12ms
38 timing idealTree:userRequests Completed in 4ms
39 silly idealTree buildDeps
40 silly fetch manifest [email protected]
41 http fetch GET 200 https://registry.npmjs.org/node-red-contrib-home-assistant-websocket 148ms (cache miss)
42 silly placeDep ROOT node-red-contrib-home-assistant-websocket@ OK for: [email protected] want: 0.59.0
43 timing idealTree:#root Completed in 153ms
44 timing idealTree:node_modules/node-red-contrib-home-assistant-websocket Completed in 1ms
45 timing idealTree:buildDeps Completed in 155ms
46 timing idealTree:fixDepFlags Completed in 0ms
47 timing idealTree Completed in 172ms
48 timing command:install Completed in 181ms
49 verbose type invalid-json
50 verbose stack FetchError: invalid json response body at https://registry.npmjs.org/node-red-contrib-home-assistant-websocket reason: Invalid response body while trying to fetch https://registry.npmjs.org/node-red-contrib-home-assistant-websocket: ENOSYS: function not implemented, link '/data/.npm/_cacache/tmp/5fb7a063' -> '/data/.npm/_cacache/content-v2/sha512/5c/6d/39a29f9624fe29b833e5a3f0c7dc3114b8613d3a6d9a560353079424535eb1bd68d551124c1d71926d968f2a89210a200a36d2c820c3a018123ffac54037'
50 verbose stack     at Response.json (/usr/local/lib/node_modules/npm/node_modules/minipass-fetch/lib/body.js:79:13)
50 verbose stack     at async RegistryFetcher.packument (/usr/local/lib/node_modules/npm/node_modules/pacote/lib/registry.js:99:25)
50 verbose stack     at async RegistryFetcher.manifest (/usr/local/lib/node_modules/npm/node_modules/pacote/lib/registry.js:124:23)
50 verbose stack     at async Arborist.[nodeFromEdge] (/usr/local/lib/node_modules/npm/node_modules/@npmcli/arborist/lib/arborist/build-ideal-tree.js:1061:19)
50 verbose stack     at async Arborist.[buildDepStep] (/usr/local/lib/node_modules/npm/node_modules/@npmcli/arborist/lib/arborist/build-ideal-tree.js:930:11)
50 verbose stack     at async Arborist.buildIdealTree (/usr/local/lib/node_modules/npm/node_modules/@npmcli/arborist/lib/arborist/build-ideal-tree.js:211:7)
50 verbose stack     at async Promise.all (index 1)
50 verbose stack     at async Arborist.reify (/usr/local/lib/node_modules/npm/node_modules/@npmcli/arborist/lib/arborist/reify.js:154:5)
50 verbose stack     at async Install.exec (/usr/local/lib/node_modules/npm/lib/commands/install.js:145:5)
50 verbose stack     at async module.exports (/usr/local/lib/node_modules/npm/lib/cli.js:78:5)
51 verbose cwd /data/.npm
52 verbose Linux 6.1.63-Unraid
53 verbose node v16.20.2
54 verbose npm  v8.19.4
55 error code FETCH_ERROR
56 error errno FETCH_ERROR
57 error invalid json response body at https://registry.npmjs.org/node-red-contrib-home-assistant-websocket reason: Invalid response body while trying to fetch https://registry.npmjs.org/node-red-contrib-home-assistant-websocket: ENOSYS: function not implemented, link '/data/.npm/_cacache/tmp/5fb7a063' -> '/data/.npm/_cacache/content-v2/sha512/5c/6d/39a29f9624fe29b833e5a3f0c7dc3114b8613d3a6d9a560353079424535eb1bd68d551124c1d71926d968f2a89210a200a36d2c820c3a018123ffac54037'
58 verbose exit 1
59 timing npm Completed in 278ms
60 verbose unfinished npm timer reify 1701626937983
61 verbose unfinished npm timer reify:loadTrees 1701626937989
62 verbose code 1
63 error A complete log of this run can be found in:
63 error     /data/.npm/_logs/2023-12-03T18_08_57_891Z-debug-0.log

 

Edited December 3, 20232 yr by stanger89

I'm using MQTT for some years now and is running fine.
I recently recognized that the appdata/MQTT directory is empty?

How is this possible, because it is still running?

Also the registery page is empty?

What do I need to do?

Edited December 17, 20232 yr by canedje2

I just found that couchdb isn't persisting its config.

 

I added CORS configuration, and it started doing what I needed it to do perfectly. But as soon as the docker had to restart (or maybe after an update? not sure actually), this bit of configuration is gone, and all config fields are back to what looks like default/initial config.

 

Can this be fixed please? This is probably a config in the dockerfile or definition of fields for Unraid. Pretty sure this is not a bug in couchdb itself.

Edited January 5, 20242 yr by thany

On 1/5/2024 at 11:41 PM, thany said:

I just found that couchdb isn't persisting its config.

 

 

Yep. This container template deletes the entire database on upgrade. Hope no-one was storing anything important.

 

appdata/couchdb directory is mapped to Container Path: /usr/local/var/lib/couchdb which appears to be unused.

 

The web config refers to "./data" which I think is "./opt/couchdb/data"?

 

What is the process to flag this app template as broken/unmaintained? Hoping no one else has to waste their afternoon on this.

 

28 minutes ago, _df said:

 

Yep. This container template deletes the entire database on upgrade. Hope no-one was storing anything important.

 

appdata/couchdb directory is mapped to Container Path: /usr/local/var/lib/couchdb which appears to be unused.

 

The web config refers to "./data" which I think is "./opt/couchdb/data"?

 

What is the process to flag this app template as broken/unmaintained? Hoping no one else has to waste their afternoon on this.

 

 

FWIW, my solution was to map the following paths.

/opt/couchdb/data <-> /mnt/user/appdata/couchdb/data
/opt/couchdb/etc/local.d <-> /mnt/user/appdata/couchdb/etc

 

@thany the second line above seems to fix the CORS config persistence for me but you should test carefully.

 

@spants FYI

 

I'm trying to install the "node-red-contrib-virtual-smart-home" palette so i can use Alexa commands in Node-RED, but that package expects node.js version > 18.2.0 to be installed (current in the docker is 16.20.2)

 

Can that be updated in the Node-RED docker?

Edited February 16, 20242 yr by jowi

Nothing? Is there a way i can update node.js myself? In a persistent way?

6 hours ago, jowi said:

Nothing? Is there a way i can update node.js myself? In a persistent way?

Change to the docker tag that has that nodejs version.

Ok, and how do i do that? 

Hello. I am running into an annoyance with updates to this container where every time the container is updated, all of the themes located in /mnt/user/appdata/nodered/@node-red-contrib-themes get wiped and I need to download the themes and add them back to this folder. Is there a way around this?

Installed NodeRed for the first time and cannot seem to get to the WebUI using either http or https and I have no idea why. There are no errors in the log. Is there some other configuration I need to make that I'm missing?
 

2 hours ago, SnugglyDino said:

Installed NodeRed for the first time and cannot seem to get to the WebUI using either http or https and I have no idea why. There are no errors in the log. Is there some other configuration I need to make that I'm missing?
 

Looks like you are giving the container it's own IP address? If so Port mappings won't do anything and it's probably listening on the default port of 1880. Also make sure you use the IP of the container not host.

 

If you want the port exposed on the host use a different network type.

On 8/26/2024 at 7:10 PM, Skylord123 said:

Looks like you are giving the container it's own IP address? If so Port mappings won't do anything and it's probably listening on the default port of 1880. Also make sure you use the IP of the container not host.

 

If you want the port exposed on the host use a different network type.

I'm confused can you please elaborate.

 

Why would changing the port to 1881 in the Port field of the Docker Setup up not change the port value from 1880 to 1881? I've done that with countless other containers and it has worked just fine. I changed it to not conflict with Nginx Proxy Manager but I really didn't need to since NPM runs on a different VLAN now that I'm thinking about it. Changing the port value back to the default 1880 in the Docker Setup unfortunately does not fix anything either (see pictures).

 

@Skylord123 Would you be able to send a picture of your setup file so I can copy it? Is there anything different?

11 hours ago, SnugglyDino said:

Why would changing the port to 1881 in the Port field of the Docker Setup up not change the port value from 1880 to 1881?

Because port mapping only applies to bridge mode, where the container shares the host's IP address. Other modes require port changes to be made in the application itself, there is no remapping because the container has total control over a unique IP instead of sharing the hosts IP.

13 hours ago, JonathanM said:

Because port mapping only applies to bridge mode, where the container shares the host's IP address. Other modes require port changes to be made in the application itself, there is no remapping because the container has total control over a unique IP instead of sharing the hosts IP.

You are right. Thank you for that clarification. Changing the network to a bridge that shares the same IP as my host system fixed the problem. How can I manually set the IP address for Node Red if I want to put it on a network that's on a different VLAN? I want it on the same network as my Home Assistant VM. You said I have to do it in the settings of the application itself.

 

Where is this setting in Node Red?

 

Is there really no way to set it from the container setup?

I had this a few times now, that CouchDB isn't persisting its settings, or at least its CORS settings. At some point, not neccesarily even after restarting the container, so perhaps after updating it only, I'm getting CORS errors in the clients "suddenly" and I have to get in there to fix it again.

 

This has happened 3 or 4 times over the past few months. So let's say this happens about once every 2 months or so - not sure that this is helpful though.

 

I'm not seeing any files appearing in the mapped config folder. That folder has 777 permissions, so it should be able to create its persistent config there, if it wants to. Otherwise I don't know where else the config goes. The databases are persisted perfectly fine, so it is capable of persisting data, generally speaking.

 

What can be done to fix this?

Edited October 7, 20241 yr by thany

yesh me too.. suggest withdrawing this template as it is not workable in current form.

 

Anyone seen this from the log?

 1738585922: Saving in-memory database to /config/data/mosquitto.db.
1738585922: Error saving in-memory database, unable to open /config/data/mosquitto.db.new for writing.
1738585922: Error: Permission denied.

My config:

6.12.14

On 10/14/2023 at 2:11 PM, camens said:

 

You just have to add COUCHDB_USER= and COUCHDB_PASSWORD= as variables in the docker setup.

 

 

1. Click on the container and click 'Edit'
2. In the Update Container window, go to the bottom and click '+ Add another Path, Port, Variable...'
3. In the Add Configuration window that pops up, change Config Type to variable, set the name and key to 'COUCHDB_USER' and the value to your username. Click Add.
4. Repeat step 3 for COUCHDB_PASSWORD.

5. Click Apply

On 2/3/2025 at 7:34 AM, orlando500 said:

Anyone seen this from the log?

 1738585922: Saving in-memory database to /config/data/mosquitto.db.
1738585922: Error saving in-memory database, unable to open /config/data/mosquitto.db.new for writing.
1738585922: Error: Permission denied.

My config:

6.12.14

Run the following commands inside the container

find /config -type d -exec chmod 775 {} \;

find /config -type f -exec chmod 664 {} \;

chown -R mosquitto:mosquitto /config

The problem is run.sh inside the container has this line chmod -R 666 /config which messed the folder permissions up.

will; this be getting an update?

17 hours ago, stackner said:

will; this be getting an update?

I switched over to the official Mosquitto container, but that didn't resolve the issue.
I realised that I had to go and reconfigure the HA MQTT integration and update the protocol version there. That resolved the issue, and I would bet I could have stayed with the spants container.

Edited June 6Jun 6 by kabadisha
Fixed a typo

I'm still on the spants MQTT container, but when I set the protocol version to 5 in home assistant it fails to connect. I think both need to be changed in order to make it work. I considered migrating to the official Mosquitto docker, but was working on other stuff today. Looks like I have some time to figure it out though based off the error from home assistant.

I see the following when it's set to 5, but nothing in the system logs to tell me exactly what's failing. Works when set to 3.1.1 though.

17 hours ago, Jurykov said:

I'm still on the spants MQTT container, but when I set the protocol version to 5 in home assistant it fails to connect. I think both need to be changed in order to make it work. I considered migrating to the official Mosquitto docker, but was working on other stuff today. Looks like I have some time to figure it out though based off the error from home assistant.

I see the following when it's set to 5, but nothing in the system logs to tell me exactly what's failing. Works when set to 3.1.1 though.

i do not beleive the SPANTS one supports version 5 is the issue.