search

munnerz / kube-plex

Scalable Plex Media Server on Kubernetes -- dispatch transcode jobs as pods on your cluster!
Apache License 2.0
1.22k stars 191 forks source link

Problem with health probes #80

Closed Krm1t closed 4 years ago

Krm1t commented 4 years ago

Hello

I'm trying to get this up and running but am running into problems with the live- and readyness probes. I keep getting Connection Refused on them.

I've already tried to disable the probes so i could connect to the container. If i then run "curl localhost:32400" or "curl localhost:32400/identity" from within the container i get connection refused on both calls.

Both of my shares are nfs shares and in the config folder the following files are created. /.bash_history /Library/Application Support/Plex Media Server/Preferences.xml

I would have expected more than only these 2 files. Also the xml file contains only the root element with a few attributes and no inner xml - Don't know if there should be more or not.

I use the following script to install kube-plex.

!/bin/bash

PLEXTOKEN=${1?Error: no plex token given}

kubectl apply -f ns-plex.yaml kubectl apply -f pvc-config-plex.yaml kubectl apply -f pvc-shared-media.yaml

helm install plex ./kube-plex/charts/kube-plex \ --namespace=plex \ --set claimToken=$PLEXTOKEN \ --set persistence.data.claimName=pvc-shared-media \ --set persistence.config.claimName=pvc-config-plex \ --set persistence.config.subPath=plex \ --set ingress.enabled=false \ --set timezone=Europe/Copenhagen

My kubernetes is setup as a single control plane as i only have 1 machine. I use calico as my network plugin. My nfs shares are located on a nas - In the unlikely event that the shares could have some influence on the issue.

I have run out of ideas for debugging this myself so i'm hoping to get some input here. Please let me know if there is any other information needed to help diagnose this.

Krm1t commented 4 years ago

For anyone having the same issue I finally figured it out. My 2 persistent volumes pointing to my NFS shares were wrong.

My nfs section was specified like this

nfs: 
  server: servername
  path: /path/to/volume

Once i changed it to the following the pod started working

nfs:
  server: servername.mydomain.local
  path: "/path/to/volume"
mountOptions:
  - rw

I'm not sure which of the 3 changes that did the trick or a combination of them all. The funny thing is that i could see files be created in the folder so it was somehow working just not 100%