Incident: gcr.io/google_containers/volume-nfs:0.8 ceased to exist

We are using mentioned image for our NFS caching of firmware to reduce the costs.
Unfortunately something happened with that image and it ceased to exist, which caused our NFS server to fail and
consequently computing clusters stopped to compile kernels.
Temporary i found alternative image, but it is not ideal, because it is not updated and it is not ours: docker.io/adnanhodzic/nfs-server-k8s
I will update the image in all clusters to the new one, but we need to find a long term solution for this, either by maintaining our own image or by finding a reliable one.
At the moment of writing work still on progress.

No files or tests are named. Start by reviewing how the NFS image is selected and deployed across the clusters, then compare the temporary image with options for a maintained replacement; done means a reliable image is chosen, deployed, and kernel compilation no longer depends on the vanished image.

Written by the indexing model from the issue text.