Closed Noah-Huppert closed 8 months ago
I'm looking into it but I'm thinking that this could be an issue with my infrastructure.
I am running Ceph and it looks like issues with your Ceph cluster can cause permission denied errors: https://access.redhat.com/solutions/7004783
Hi, what did you find? Are you still having an issue with this?
Hi, what did you find? Are you still having an issue with this?
I found that my Ceph OSDs were running out of memory, crashing, and causing the Ceph cluster to thrash and restart loop. This then caused the Emby error because it couldn't access files. I never found the root cause but this mailing list post has some good ideas I was going to try. I ended up resorting to using plain old Kubernetes PVCs and placement affinity instead of a distributed storage cluster.
Thanks for the info.
Recently while playing the last 3 episodes of a TV show Emby has crashed 10-20 minutes in. In the logs I see a
NullReferenceException
, the first method in the stack trace isEmby.Api.Images.ImageService.GetImage
. After about 5 minutes the server recovers and I can start viewing media again.I am running Emby on Kubernetes via the
emby/embyserver:4.8.0.39
image. The media and config directories are provided by Rook Ceph. I am viewing media via the Nvidia Shield Emby app.Full Logs