Does anyone else use inject for a specific time execution? Reading all the posts on this issue in various forums, this is UI only and doesn’t affect operation. I wonder if I need to test that again and make sure it is not something unrelated to the UI issue.
Can you give an example node and I’ll test it on mine.
OK, I just test one of my Inject Nodes scheduled for 10am… it worked with no issues.
I did notice that when I accessed node-red through port 1800, instead of going through HA, that node-red showed me a list of errors/warning in the upper right corner. I don’t get this when going through HA. I have been working to resolve these issues to see if it resolves my problems.
Have you checked this to see if it’s telling you what’s wrong on your end?
I noticed the UI was different and had a bunch of errors. I didn’t dig into those. I’ll spend some time working through those and see what comes of it.
OK, good news… once I resolved all the errors and actually hit deploy to resolve the last one, node-red automatically rebooted the service and now everything is working through HA again!
1 Like
Here is what I did:
I resolved this by going to node-red directly through the 1800 port (http://node-red-ip:1800). Once into node-red, in the upper right corner, node-red showed a yellow warning sign with X number of issues. Clicking this shows you all nodes with errors or references to Home Assistant entities that are missing/wrong.
I went through all these issue and resolved them. When I got the the last one and hit ‘deploy’, node-red automatically restarted the node-red service. When it came back up, I check Node-red through HA and everything is working again!
Either they changed something in the new version that doesn’t like nodes with errors, or this a bug.
1 Like
Nevermind… it worked for about 30 mins then went right back to the same issue. ugh!
Had the same issue…
local node red fixed all the issues top right corner.
Then restart HA and restart node-red.
local node red IP debug and inject is working.
Homeassistant local IP in node red not working…
Even when I am able to click a Timestamp, most of the time nothing happens. I hooked a Debug to a Timestamp, clicked the inject and got nothing in Debug. This really needs to be fixed.
Hmmm… not seeing any of that here. I’m using the HA Add-On and the latest versions of everything.
At least for me this was fixed by accessing NodeRed directly through the browser rather than launching it from within HA. To do that go to http://:1880/ you will be prompted for a user/pass, enter your HA details. At that point everything started working again.
Worth noting looks like this has been happening to people for a very long time, I was able to find multiple threads discussing this issue.
This isn’t the first or even the second time this has happened to me. It’s been years since the last time this affected me, I’ve been using 1880 probably since the first time it happened.
This time, all access points took several minutes to load and inject/debugs were grayed out when they did. This isn’t the first time that, that has happened either. I’m convinced that it is some type of corruption during the install, likely in the network/auth stack.
It’s always been an access issue, NR seems to run fine as far as executing automations. In the past I would back up my flows and remove and reinstall the addon. This last time though I restored a backup. Only because I since enabled store to file inside NR, so I have data stored in there that is not backed up with the flows.
I’ve never needed to wait for the next update for this to be fixed, it was always just a process of removing it and reinstalling the addon.
That works fine at home, but how do I access that via Nabu Casa? I can’t just add the :1880 to the end of the Nabu address.
Why not? add it after .com .org or whatever the end domain is.
www.example.nabucasa.com:1880
Edit: I assume nabu casa works like other https components, you may need to forward the request from the nabu casa domain to 1880 in your firewall. Iirc that’s how I did it when I used duckdns.
Toggling “Using the Home Assistant Add-on” (1) off and deploy and then (2) on and deploy solved it for me. I guess it refreshes some credentials or something.
I have this problem as well, and the first thing I did was roll back to what I’m on now:
* Core2024.9.3
* Supervisor2024.10.3
* Operating System13.1
* Frontend20240909.1
* NodeRED Add-on 18.0.5
* node-red-contrib-home-assistant-websocket 0.65.1
I’m going to log my actions as I backup and upgrade now.
Bumping up to 2024.10.4 from 2024.9.3 with a full backup prior…success.
Bumping HAOS to 13.2 from 13.1…success.
Bumping NodeRED 18.0.5 to 18.1.1…success.
Bumping Configuration Node from 0.65.1 to 0.74.2…success.
Everything is working up to this point with no troubleshooting needed.
Updating all HA Nodes…
…aside from a host of issues with stragglers in NodeRED, the problem appears to be gone.
Current Setup:
* Core2024.10.4
* Supervisor2024.10.3
* Operating System13.2
* Frontend20241002.4