Kubernetes: mounting nfs volumes in pods is broken

[sjenning]

https://github.com/kubernetes/kubernetes/pull/35652 broke mounting nfs volumes in pods:

apiVersion: v1
kind: Pod
metadata:
  name: demo
spec:
  containers:
  - image: fedora:24
    name: fedora
    imagePullPolicy: Always
    command:
    - /usr/bin/sleep
    - "3600"
    volumeMounts:
    - name: myvol
      mountPath: /mnt
  volumes:
  - name: myvol
    nfs:
      path: /srv/share
      server: localhost
  terminationGracePeriodSeconds: 5

kubectl create -f nfspod.yaml

Stuck in ContainerCreating

kubectl describe pod demo

Events:
  FirstSeen LastSeen    Count   From            SubObjectPath   Type        Reason      Message
  --------- --------    -----   ----            -------------   --------    ------      -------
  14s       14s     1   {default-scheduler }            Normal      Scheduled   Successfully assigned demo to 127.0.0.1
  14s       6s      5   {kubelet 127.0.0.1}         Warning     FailedMount MountVolume.SetUp failed for volume "kubernetes.io/nfs/3cc8bac7-9c77-11e6-ae11-fa163e815edd-myvol" (spec.Name: "myvol") pod "3cc8bac7-9c77-11e6-ae11-fa163e815edd" (UID: "3cc8bac7-9c77-11e6-ae11-fa163e815edd") with: mount failed: exit status 32
Mounting command: mount
Mounting arguments: /localhost:/srv/share /var/lib/kubelet/pods/3cc8bac7-9c77-11e6-ae11-fa163e815edd/volumes/kubernetes.io~nfs/myvol nfs []
Output: mount.nfs: Failed to resolve server /localhost: Name or service not known

Note the leading / on the nfs server name.

@vishh @derekwaynecarr

[vishh]

https://github.com/kubernetes/kubernetes/pull/35349/commits/7fd03c4b6e132347cfa591abb263b92a53d004b1 should fix this. Sorry for the trouble.

[sjenning]

@vishh ah yes, I had a PR with that fix ready to go but you beat me to it. Minus the whole else case there since the if before it ends in a return. Thanks! I'll close when it merged

[HJFinch]

Hey there,
the NFS Mount still seams unable to resolve service host names (service.namespace) on GCE.
Is this issue fixed and merged or does it still take some time for it to be fixed on GCE?

MountVolume.SetUp failed for volume "kubernetes.io/nfs/***" (spec.Name: "***") pod "***" (UID: "***") with: mount failed: exit status 32 Mounting command: mount Mounting arguments: nfs-server:/exports/jenkins-master /var/lib/kubelet/pods/***/volumes/kubernetes.io~nfs/jenkins-master nfs [] Output: mount.nfs: Failed to resolve server nfs-server.storage: Name or service not known

The DNS seems to work properly as running containers are able to resolve the name. Also using the IP works as well but is no solution.