informatics-lab / s3-fuse-flex-volume

A Kubernetes flex volume driver for a custom s3 fuse adapter
BSD 3-Clause "New" or "Revised" License
31 stars 11 forks source link

Installation not working #3

Closed h4gen closed 5 years ago

h4gen commented 5 years ago

Hey! Sadly I am not able to get your chart running on my cluster. This is the status of the pods:

s3-fuse-flex-volume-deployer-69w7w                         0/1       Init:CrashLoopBackOff   6          7m
s3-fuse-flex-volume-deployer-9gp5n                         0/1       Init:CrashLoopBackOff   6          7m

Pod description yields the following:

Name:           s3-fuse-flex-volume-deployer-69w7w
Namespace:      kube-system
Node:           gke-idalab-cluster-default-pool-90a688df-fcf3/10.132.0.3
Start Time:     Thu, 01 Nov 2018 15:57:58 -0400
Labels:         app=s3-fuse-flex-volume-deployer
                controller-revision-hash=2342902337
                pod-template-generation=1
Annotations:    <none>
Status:         Pending
IP:             10.16.0.14
Controlled By:  DaemonSet/s3-fuse-flex-volume-deployer
Init Containers:
  install-host-deps:
    Container ID:  docker://80ed41a4237e622a063c30f20991ae57c3a2fabdf37f5e773e4654e600e32f53
    Image:         busybox:latest
    Image ID:      docker-pullable://busybox@sha256:915f390a8912e16d4beb8689720a17348f3f6d1a7b659697df850ab625ea29d5
    Port:          <none>
    Host Port:     <none>
    Command:
      sh
      -c
      chroot /rootfs apt-get update && chroot /rootfs apt-get install -y fuse python3-pip && chroot /rootfs pip3 install git+git://github.com/met-office-lab/pysssix.git@big_cache && chroot /rootfs curl -L -o /usr/bin/goofys http://bit.ly/goofys-latest && chroot /rootfs chmod +x /usr/bin/goofys
    State:          Waiting
      Reason:       CrashLoopBackOff
    Last State:     Terminated
      Reason:       Error
      Exit Code:    127
      Started:      Thu, 01 Nov 2018 16:03:30 -0400
      Finished:     Thu, 01 Nov 2018 16:03:30 -0400
    Ready:          False
    Restart Count:  6
    Environment:    <none>
    Mounts:
      /rootfs from rootfs (rw)
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-jmz5r (ro)
  s3-flex-volume-drivers-deploy:
    Container ID:
    Image:          informaticslab/s3-fuse-flex-volume:latest
    Image ID:
    Port:           <none>
    Host Port:      <none>
    State:          Waiting
      Reason:       PodInitializing
    Ready:          False
    Restart Count:  0
    Environment:    <none>
    Mounts:
      /flexmnt from flexvolume-plugindir (rw)
      /hostbin from flexvolume-usrlocalbin (rw)
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-jmz5r (ro)
Containers:
  pause:
    Container ID:
    Image:          gcr.io/google-containers/pause
    Image ID:
    Port:           <none>
    Host Port:      <none>
    State:          Waiting
      Reason:       PodInitializing
    Ready:          False
    Restart Count:  0
    Environment:    <none>
    Mounts:
      /var/run/secrets/kubernetes.io/serviceaccount from default-token-jmz5r (ro)
Conditions:
  Type           Status
  Initialized    False
  Ready          False
  PodScheduled   True
Volumes:
  flexvolume-plugindir:
    Type:          HostPath (bare host directory volume)
    Path:          /usr/libexec/kubernetes/kubelet-plugins/volume/exec/
    HostPathType:
  flexvolume-usrlocalbin:
    Type:          HostPath (bare host directory volume)
    Path:          /usr/local/bin
    HostPathType:
  rootfs:
    Type:          HostPath (bare host directory volume)
    Path:          /
    HostPathType:  Directory
  default-token-jmz5r:
    Type:        Secret (a volume populated by a Secret)
    SecretName:  default-token-jmz5r
    Optional:    false
QoS Class:       BestEffort
Node-Selectors:  <none>
Tolerations:     node.kubernetes.io/disk-pressure:NoSchedule
                 node.kubernetes.io/memory-pressure:NoSchedule
                 node.kubernetes.io/not-ready:NoExecute
                 node.kubernetes.io/unreachable:NoExecute
Events:
  Type     Reason                 Age               From                                                    Message
  ----     ------                 ----              ----                                                    -------
  Normal   SuccessfulMountVolume  8m                kubelet, gke-idalab-cluster-default-pool-90a688df-fcf3  MountVolume.SetUp succeeded for volume "flexvolume-plugindir"
  Normal   SuccessfulMountVolume  8m                kubelet, gke-idalab-cluster-default-pool-90a688df-fcf3  MountVolume.SetUp succeeded for volume "rootfs"
  Normal   SuccessfulMountVolume  8m                kubelet, gke-idalab-cluster-default-pool-90a688df-fcf3  MountVolume.SetUp succeeded for volume "flexvolume-usrlocalbin"
  Normal   SuccessfulMountVolume  8m                kubelet, gke-idalab-cluster-default-pool-90a688df-fcf3  MountVolume.SetUp succeeded for volume "default-token-jmz5r"
  Normal   Pulling                7m (x4 over 8m)   kubelet, gke-idalab-cluster-default-pool-90a688df-fcf3  pulling image "busybox:latest"
  Normal   Pulled                 7m (x4 over 8m)   kubelet, gke-idalab-cluster-default-pool-90a688df-fcf3  Successfully pulled image "busybox:latest"
  Normal   Created                7m (x4 over 8m)   kubelet, gke-idalab-cluster-default-pool-90a688df-fcf3  Created container
  Normal   Started                7m (x4 over 8m)   kubelet, gke-idalab-cluster-default-pool-90a688df-fcf3  Started container
  Warning  BackOff                3m (x24 over 8m)  kubelet, gke-idalab-cluster-default-pool-90a688df-fcf3  Back-off restarting failed container

And thats the respond to kubectl logs

Error from server (BadRequest): container "pause" in pod "s3-fuse-flex-volume-deployer-9gp5n" is waiting to start: PodInitializing

Any idea whats going wrong?

jacobtomlinson commented 5 years ago

Can you run kubectl logs -c install-host-deps. The work in this chart is done in the init container.

h4gen commented 5 years ago

The output of

kubectl logs --namespace kube-system s3-fuse-flex-volume-deployer-srf7j -c install-host-deps

is

chroot: can't execute 'apt-get': No such file or directory

for every pod!

jacobtomlinson commented 5 years ago

This plugin assumes you are using a Debian based distribution on your kubernetes cluster. From that error I'm guessing you are not.

I would be keen to make this more versatile and support more distributions but I don't have the bandwidth to do that right now.

jacobtomlinson commented 5 years ago

Closing due to lack of response.