Did you also try http_node & http_static blank?
Otherwise just deinstall and reinstall. I had that happen sometimes with other addons that suddenly work correctly after reinstallation.
Why’s the server running on port 46836? Shouldn’t it run on 1880?
If you go to the IP address with port 46836 can you get to the editor that way?
Update- After checking the release notes it appears the addon version of Node Red is supposed to run on port 46836 and then use ingress to access Home Assistant, so not sure what could be going wrong here. There was some posts about people having issues accessing the editor on the Node Red Add On thread starting here after that change was made -
No, this is not the first install of the Node Red addon.
I have been reinstall my host computer and restored a backup/snapshot where I hade Node Red installed.
Node Red have been working before.
I tried to uninstall and reinstall the Node Red addon. But no different.
I also tried different configuration of http_node & http_static. But no different.
@mwav3 that is strange?
I run my HA supervised on a computer with ip-address 10.0.1.3 and logg in to HA with 10.0.1.3:8123.
I can try http://127.0.0.1:46836.
But of course would I like to change to port 1880 and the ip-address
Have you tried another browser or computer? If the problem is still the same you could try a full remove and reinstall
Export all of your flows to a file. Then remove nodered. Then inside you config folder rename the nodered folder. This contains all the configuration information for node red and it persists over un installs
127.0.0.1 is the “loopback” address, and just points back to the machine you’re currently on. It might work if you were on the actual machine home assistant is installed on, but wouldn’t from a different one. With ingress, Node Red shold show up as an option on the side menu and you should be able to pull the editor up from anywhere you see your home assistant page
If you have 1880 listed in the addon configuration settings in the supervisor tab, it should put the editor on port 1880 as well. The log would still say 46836 but home assistant would put up the page on 1880 in the background. So if ip is 10.0.1.3 it should pull up the editor at 10.0.1.3:1880
That should all happen seamlessly to the user without them having to do anything else besides set the config properly in the addon settings on the supervisor tab. If something went wrong, best to follow the advice from mikefila above to export flows, rename folder, and reinstall.
If you can’t get access to the editor the flows are stored in a flows.json file and can be restored from there. That file is in the config/node-red directory.
What I would do first is uninstall the add-on. Then rename the old node red directory to something like node-redold. Reinstall the addon- at that point you should have a clean slate.
Then in the editor, go to import. You can use the file editor to navigate to node-redold and open the flows.json file. Copy the entire contents of the flows.json file and then paste it into the clipboard part of the import menu in your empty new node red instance.
You’ll probably have to reinstall any palettes and setup your credentials again for any flows.
I’m assuming based on your logs and issue there must have been some corruption in the old config, or it wasn’t compatible with your new install. Maybe you went from an rpi to another system with a different architecture and its creating a conflict. The more you try and copy over from the old config, the more likely you’ll carry the problem over. Hopefully importing the flows doesn’t just keep crashing the editor.
Okay @mwav3 I have tried to reinstall the NodeRed addon and renamed the NodeRed folder and reinstall the addon again.
But with the same result
Only a blank page when I try to open the NodeRed UI with a Bad Gateway error!
This is how my NodeRed logger looks like after the addon has been reinstalled.
[s6-init] making user provided files available at /var/run/s6/etc...exited 0.
[s6-init] ensuring user provided files have correct perms...exited 0.
[fix-attrs.d] applying ownership & permissions fixes...
[fix-attrs.d] done.
[cont-init.d] executing container initialization scripts...
[cont-init.d] 00-banner.sh: executing...
-----------------------------------------------------------
Add-on: Node-RED
Flow-based programming for the Internet of Things
-----------------------------------------------------------
Add-on version: 10.1.0
You are running the latest version of this add-on.
System: Debian GNU/Linux 11 (bullseye) (amd64 / qemux86-64)
Home Assistant Core: 2021.11.3
Home Assistant Supervisor: 2021.10.8
-----------------------------------------------------------
Please, share the above information when looking for help
or support in, e.g., GitHub, forums or the Discord chat.
-----------------------------------------------------------
[cont-init.d] 00-banner.sh: exited 0.
[cont-init.d] 01-log-level.sh: executing...
[cont-init.d] 01-log-level.sh: exited 0.
[cont-init.d] customizations.sh: executing...
[cont-init.d] customizations.sh: exited 0.
[cont-init.d] nginx.sh: executing...
[cont-init.d] nginx.sh: exited 0.
[cont-init.d] node-red.sh: executing...
patching file nodes/ui_base.html
Hunk #1 succeeded at 1162 (offset 631 lines).
[cont-init.d] node-red.sh: exited 0.
[cont-init.d] done.
[services.d] starting services
[services.d] done.
[20:26:44] INFO: Starting Node-RED...
> start
> node $NODE_OPTIONS node_modules/node-red/red.js "--settings" "/etc/node-red/config.js"
15 Nov 20:26:46 - [info]
Welcome to Node-RED
===================
15 Nov 20:26:46 - [info] Node-RED version: v2.1.3
15 Nov 20:26:46 - [info] Node.js version: v14.18.1
15 Nov 20:26:46 - [info] Linux 5.10.0-9-amd64 x64 LE
15 Nov 20:26:46 - [info] Loading palette nodes
15 Nov 20:26:48 - [info] Dashboard version 3.1.1 started at /endpoint/ui
/bin/sh: /opt/node_modules/node-red-node-pi-gpio/testgpio.py: not found
15 Nov 20:26:49 - [warn] rpi-gpio : Raspberry Pi specific node set inactive
15 Nov 20:26:49 - [info] Settings file : /etc/node-red/config.js
15 Nov 20:26:49 - [info] Context store : 'default' [module=memory]
15 Nov 20:26:49 - [info] User directory : /config/node-red/
15 Nov 20:26:49 - [warn] Projects disabled : editorTheme.projects.enabled=false
15 Nov 20:26:49 - [info] Flows file : /config/node-red/flows.json
15 Nov 20:26:49 - [info] Server now running at http://127.0.0.1:46836/
15 Nov 20:26:49 - [info] Starting flows
15 Nov 20:26:49 - [info] Started flows
[20:26:49] INFO: Starting NGinx...
15 Nov 20:26:54 - [info] [server:Home Assistant] Connecting to undefined
15 Nov 20:26:54 - [info] [server:Home Assistant] Connected to undefined
I’m not sure if order matters but your config looks a little different then the example one, you can try and format the config more like the documentation example here:
That documentation also notes specifically under known issues and limitations:
If you cannot access HTTP nodes or Node-RED Dashboard, please check if you have enabled direct access mode by setting a port number in “Network” configuration section of the add-on.
If you cannot access HTTP nodes or Node-RED Dashboard, please check if you URL starts with /endpoint/, or else Home Assistant authentication will kick in.
Based on the advice above, try removing the port number on the network configuration tab to disable the “direct access” and only access the dashboard through Home Assistant ingress (ie just the Node Red tab in Home Assistant).
You also can try the config this poster suggested
I wish I had a better answer that I knew would fix this for you, but I just run Node Red in docker and not the addon, so I’m not familiar with all the quirks and finer details of getting the addon to work other then what we’ve tried. Unlike a traditional Node Red install (ie docker, snap, NPM) that just runs on port 1880, the Node Red addon appears to route the connection to the supervisor and all over the place through a bunch of different ports and uses ingress so a lot can go wrong. Hopefully someone else can chime in if they have any other ideas.
If all else fails you can just give up on the addon and run Node Red directly on port 1880 outside of Home Assistant, instructions to do that are below
What error do you get when using direct access in lieu of going through HA eg. Your HA address with :1880 rather than :8123 If your using https you will have to make a security exception in the browser to display the page.
Have you tried changing to a different port on the config page?
I wonder if the work around from an old HACS problem would help. Force disabling the cache for the page.
@Mikefila, I’m using https. How do I do the security exception?
Do I have to do an execption on each browser I’m using?
I.e I log on to the system from different computers.
I tried to remove the port (1880) as @mwav3 suggested. But no change
When you go to the page in a browser there will be a warning there will also be a link or button that will give you further options. You can try another port like 9025 and then use that at the end of your ip instead of 8123
