2025-08-20, 12:01 PM
(This post was last modified: 2025-08-20, 12:02 PM by SpiderUnderUrBed. Edited 1 time in total.)
https://pastebin.com/Qu1VPsMf
jellyfin is the only service in my k8s cluster which I cannot curl, its also the only thing that runs on another node. How do i fix this so i can access jellyfin through my ingress?
I am able to ping it from a container, the main issue is 8096 I think, that and/or pod-to-pod connectivity between nodes as I think it worked when jellyfin was on the same node as traefik? but there is also some weird jellyfin stuff happening. If it listens to 0.0.0.0 by default then I am really confused as to whats happening, if its only listening to localhost that makes more sense, here is some more information (my traefik), also I installed metallb to try and fix this issue, and I always had cilium running as my CNI
traefik:
https://pastebin.com/7ANL8T8L
svc, pods, ingresses, etc:
https://pastebin.com/uumFPiDu
Also since I recently installed metallb to try and fix this issue, the ip metallb got for my ingress isnt working with the curl commands too, well with everything except jellyfin, and I did set the jellyfin domain env variable correctly.
Code:spiderunderurbed@raspberrypi:~/k8s $ curl -H "Host: jellyfin.spidershomelab.xyz" http://192.168.0.22:80
^C
spiderunderurbed@raspberrypi:~/k8s $ curl -k -H "Host: jellyfin.spidershomelab.xyz" https://192.168.0.22:80
404 page not found
jellyfin is the only service in my k8s cluster which I cannot curl, its also the only thing that runs on another node. How do i fix this so i can access jellyfin through my ingress?
Code:spiderunderurbed@raspberrypi:~/k8s $ kubectl get ingress
NAME CLASS HOSTS ADDRESS PORTS AGE
gitea-ingress traefik gitea.spidershomelab.xyz 192.168.0.22,192.168.0.36 80 6d
glance-ingress traefik glance.spidershomelab.xyz 192.168.0.22,192.168.0.36 80 6d
gotify-ingress traefik gotify.spidershomelab.xyz 192.168.0.22,192.168.0.36 80 6d
jellyfin-ingress traefik jellyfin.spidershomelab.xyz 80 3m4s
nextcloud-ingress traefik nextcloud.spidershomelab.xyz 192.168.0.22,192.168.0.36 80 6d
uptime-kuma-ingress traefik uptime-kuma.spidershomelab.xyz 192.168.0.22,192.168.0.36 80 6d
vaultwarden-ingress traefik vaultwarden.spidershomelab.xyz 192.168.0.22,192.168.0.36 80 6d
wordpress-ingress traefik spidershomelab.xyz 192.168.0.22,192.168.0.36 80 6d
spiderunderurbed@raspberrypi:~/k8s $
I am able to ping it from a container, the main issue is 8096 I think, that and/or pod-to-pod connectivity between nodes as I think it worked when jellyfin was on the same node as traefik? but there is also some weird jellyfin stuff happening. If it listens to 0.0.0.0 by default then I am really confused as to whats happening, if its only listening to localhost that makes more sense, here is some more information (my traefik), also I installed metallb to try and fix this issue, and I always had cilium running as my CNI
traefik:
https://pastebin.com/7ANL8T8L
svc, pods, ingresses, etc:
https://pastebin.com/uumFPiDu
Also since I recently installed metallb to try and fix this issue, the ip metallb got for my ingress isnt working with the curl commands too, well with everything except jellyfin, and I did set the jellyfin domain env variable correctly.