Unable to access Tools/Docker & Tools/Tasks
When I try to access Tools/Docker I get: Failed to fetch Docker containers
When I try to access Tools/Taks I get: Application error: a server-side exception has occurred while loading homarr.star-one.co.uk (see the server logs for more information). Digest: 3054471037
Both used to work but having not used them for some time I don't know when they stopped working. When I check the logs in debug, I see messages similar in bot like the following:
2025-10-29T07:41:31.926Z info: tRPC request from unknown by user 'ferd (e<redacted>k)'
2025-10-29T07:41:38.579Z debug: Rewriting origin from http://homarr.mydomain.com:3000 to http://homarr.mydomain.com
2025-10-29T07:43:45.684Z debug: Rewriting origin from http://homarr.mydomain.com:3000 to http://homarr.mydomain.com
I mention these because those are not the URLs used to connect to Homarr - the domains are also redacted. I'm not sure why it uses port 3000
I use HAProxy on my OPNsense firewall to proxy the connection https://homarr.mydomain.com to my Homarr container http://maggie.mydomain.com:7575 and as per all my other proxied services I have enabled the forwarding options.
So I think that it's not reading the forwarded header from HAProxy.
Any suggestions?
Homarr v1.43.0
Solution:Jump to solution
I rebuilt the container to more easily see an issue with Redis, so far it's fine and even the Tasks is showing. I'll see see if it stops working again in the coming hours.
22 Replies
Thank you for submitting a support request.
Depending on the volume of requests, our team should get in contact with you shortly.
⚠️ Please include the following details in your post or we may reject your request without further comment: - Log (See https://homarr.dev/docs/community/faq#how-do-i-open-the-console--log) - Operating system (Unraid, TrueNAS, Ubuntu, ...) - Exact Homarr version (eg. 0.15.0, not latest) - Configuration (eg. docker-compose, screenshot or similar. Use ``your-text`` to format) - Other relevant information (eg. your devices, your browser, ...)
Frequently Asked Questions | Homarr documentation
Can I install Homarr on a Raspberry Pi?
What did you set the forwarded header to?
There's no option to set it to anything, the screen shot shows all there is, and this works for all my other proxied connections as well e.g. Home Assistant, Guacamole, SABnzbd
I just remembered I can check the HAProxy settings configured by OPNsense, and for Homarr it's this:
backend Homarr
# health checking is DISABLED
mode http
balance source
# stickiness
stick-table type ip size 50k expire 30m
stick on src timeout server 30m http-reuse never option forwarded option forwardfor server Homarr_Maggie maggie.mydomain.com:7575 resolve-prefer ipv4 But I can't change them other than through the main GUI
stick on src timeout server 30m http-reuse never option forwarded option forwardfor server Homarr_Maggie maggie.mydomain.com:7575 resolve-prefer ipv4 But I can't change them other than through the main GUI
I'm a bit confused by all the domains and hosts. So Homarr is supposed to run on maggie.mydomain.com, correct?
But you're saying that doesn't work because it doesn't process the forwarded header correctly?
Yes, the Homarr container runs on a machine called maggie
Everything works except these two functions Tasks + Docker, I can see my home pages, edit them, manage integrations etc etc
I'm only guessing about the header because of what was in the logs
And you're exposing using a standard port, such as 443 or 80?
3000 in the logs is expected, we use that as an internal port for Homarr
As long as you correctly map ports and forward using a reverse proxy or tunnel, it should work just fine.
The container is expose via port 7575
This is all the logging when I try to visit both menu items: https://pastebin.com/vq61HQLK
Pastebin
Homarr Log - Pastebin.com
Pastebin.com is the number one paste tool since 2002. Pastebin is a website where you can store text online for a set period of time.
And is port 7575 in the forwarded header?
And can you check that your reverse proxy has enabled (or allows) WebSocket?
I wouldn't have though so, and now you explain the 3000 thing I don't think it's the header forward, or I suppose other stuff would fail
I've not had to deal with websockets and HAProxy for many years, the last time it wasn't working too well.
Ok. Can you please post full logs? The error you saw should also log something in the server, but your pastebin doesn't contain any errors
Full logs from where? I went to Tools/Logs, changed it to debug, then on another tab visited the two items
How are you running Homarr? I need the full logs, not from the Web UI
E.g. docker container logs
Ah ok, let me check those....
I followed the log then visited the two items like before:
⨯ Error [TRPCError]: Cannot read properties of undefined (reading 'online_cpus')
at <unknown> (.next/server/chunks/8950.js:6:123819)
at <unknown> (.next/server/chunks/8950.js:6:123922)
at async l (.next/server/chunks/8950.js:6:124110)
at async Object.requestAsync (.next/server/chunks/8950.js:6:123049)
at async Object.requestAsync (.next/server/chunks/8950.js:6:176330)
at async g (.next/server/chunks/8950.js:5:23890)
at async Object.getCachedOrUpdatedDataAsync (.next/server/chunks/8950.js:5:24358)
at async (.next/server/chunks/8950.js:135:40160) {
code: 'INTERNAL_SERVER_ERROR',
digest: '3575254010',
[cause]: TypeError: Cannot read properties of undefined (reading 'online_cpus')
at <unknown> (.next/server/chunks/8950.js:6:123819)
at <unknown> (.next/server/chunks/8950.js:6:123922)
at async l (.next/server/chunks/8950.js:6:124110)
at async Object.requestAsync (.next/server/chunks/8950.js:6:123049)
at async Object.requestAsync (.next/server/chunks/8950.js:6:176330)
at async g (.next/server/chunks/8950.js:5:23890)
at async Object.getCachedOrUpdatedDataAsync (.next/server/chunks/8950.js:5:24358)
at async (.next/server/chunks/8950.js:135:40160)
}
2025-10-29T13:40:31.855Z warn: Received message on unknown channel channel='pubSub:logging'
⨯ Error [TRPCError]: fetch failed
at a.from (.next/server/chunks/2032.js:90:47353)
at <unknown> (.next/server/chunks/2032.js:90:50773) {
code: 'INTERNAL_SERVER_ERROR',
digest: '3054471037',
[cause]: Error [TRPCClientError]: fetch failed
at a.from (.next/server/chunks/2032.js:90:47353)
at <unknown> (.next/server/chunks/2032.js:90:50773) {
shape: undefined,
data: undefined,
meta: undefined,
[cause]: TypeError: fetch failed
at async D (.next/server/chunks/2032.js:90:49199) {
[cause]: [AggregateError: ] { code: 'ECONNREFUSED' }
}
}
}
Well there seem to be quite a few errors...
@Meierschlumpf I think we have multiple issues here. Could be misconfiguration or bugs... What do you think?
Would the entire log from the container be useful? I can DM the Pastebin link rather than put it here in public, just in case it has any sensitive info.
Maybe, please let us check this one first. Please wait for a reply from us
Sure
The forwarded header including the port 3000 can be ignored. This is some internal stuff to calculate the correct redirect uri for sso auth. (It has port 3000 because it is behind the internal nginx proxy)
Docker seems to be a problem related to stats where a property is not available, if you are on the latest version, create an issue on Github otherwise update and check if it works again
Lastly tasks seems unable to connect to redis. Check your logs if it crashed/crashes and restart the container. If it doesn't work, we'll dig deeper
Regarding, sorry should have said earlier, but I use Podman on a Fedora server, but this is my build commands:
podman pod create --name Homarr --publish 7575:7575/tcp --hostname=homarr.mydomain.com
podman create --name homarr --pod Homarr --user root --no-healthcheck --volume /root/homarr_home/appdata:/appdata --volume /var/run/podman/podman.sock:/var/run/docker.sock --env LOG_LEVEL=warn --env SECRET_ENCRYPTION_KEY=<REDACTED> --env TZ=Europe/Luxembourg ghcr.io/homarr-labs/homarr:latest
So I don't really care if that's not working, I just added it because it was showing those port 3000 messages so I thought it might be related. It's probably never worked.
I needed to get to the Tasks because I wanted to check it against another problem I'm having with the Tdarr widget.
Regarding Redis (now Valkey under Fedora), where does that come from because it's only used by the local Nextcloud, not a container, which as far as I am aware working just fine.There is a redis instance in the homarr container, feel free to open a new issue with details regarding tdarr
Solution
I rebuilt the container to more easily see an issue with Redis, so far it's fine and even the Tasks is showing. I'll see see if it stops working again in the coming hours.
FYI, no further issues since yesterday, the Tasks view is now fully working and no sign of any Redis messages. Thank-you for help, I will put together a new issue for the Tdarr widget.