Azure / ACS

Azure Container Service - Bug Tracker + Announcements
65 stars 27 forks source link

Standard NC_6 VM shows 0 GPUs #116

Open DaveVoyles opened 6 years ago

DaveVoyles commented 6 years ago

I've spun up a Kubernetes cluster with this command:

az acs create 
--agent-vm-size Standard_NC6 
--resource-group dv-k8-rg 
--name dv-acs 
--orchestrator-type Kubernetes 
--agent-count 1 
 --location eastus 
 --generate-ssh-k

When I describe the nodes, it returns:

DaveVoyles@minint-55r162g  ~  kubectl describe node k8s-agent-4ef87f69-0
Name:               k8s-agent-4ef87f69-0
Roles:              agent
Labels:             agentpool=agent
                    beta.kubernetes.io/arch=amd64
                    beta.kubernetes.io/instance-type=Standard_NC6
                    beta.kubernetes.io/os=linux
                    failure-domain.beta.kubernetes.io/region=eastus
                    failure-domain.beta.kubernetes.io/zone=0
                    kubernetes.io/hostname=k8s-agent-4ef87f69-0
                    kubernetes.io/role=agent
Annotations:        node.alpha.kubernetes.io/ttl=0
                    volumes.kubernetes.io/controller-managed-attach-detach=true
CreationTimestamp:  Fri, 27 Apr 2018 09:18:44 -0700
Taints:             <none>
Unschedulable:      false
Conditions:
  Type                 Status  LastHeartbeatTime                 LastTransitionTime                Reason                       Message
  ----                 ------  -----------------                 ------------------                ------                       -------
  NetworkUnavailable   False   Fri, 27 Apr 2018 09:19:54 -0700   Fri, 27 Apr 2018 09:19:54 -0700   RouteCreated                 RouteController created a route
  OutOfDisk            False   Fri, 27 Apr 2018 09:56:22 -0700   Fri, 27 Apr 2018 09:18:44 -0700   KubeletHasSufficientDisk     kubelet has sufficient disk space available
  MemoryPressure       False   Fri, 27 Apr 2018 09:56:22 -0700   Fri, 27 Apr 2018 09:18:44 -0700   KubeletHasSufficientMemory   kubelet has sufficient memory available
  DiskPressure         False   Fri, 27 Apr 2018 09:56:22 -0700   Fri, 27 Apr 2018 09:18:44 -0700   KubeletHasNoDiskPressure     kubelet has no disk pressure
  Ready                True    Fri, 27 Apr 2018 09:56:22 -0700   Fri, 27 Apr 2018 09:19:20 -0700   KubeletReady                 kubelet is posting ready status
Addresses:
  InternalIP:  10.240.0.4
  Hostname:    k8s-agent-4ef87f69-0
Capacity:
 alpha.kubernetes.io/nvidia-gpu:  0
 cpu:                             6
 memory:                          57702220Ki
 pods:                            110
Allocatable:
 alpha.kubernetes.io/nvidia-gpu:  0
 cpu:                             6
 memory:                          57599820Ki
 pods:                            110
System Info:
 Machine ID:                 32e48d371198e8420c53b0a1fa37e94d
 System UUID:                046CA44D-B554-9A47-B488-06E9EA45C5AF
 Boot ID:                    6e74f9a1-c3dc-45b4-81e2-0898317ef53b
 Kernel Version:             4.4.0-121-generic
 OS Image:                   Debian GNU/Linux 8 (jessie)
 Operating System:           linux
 Architecture:               amd64
 Container Runtime Version:  docker://1.12.6
 Kubelet Version:            v1.7.7
 Kube-Proxy Version:         v1.7.7
PodCIDR:                     10.244.1.0/24
ExternalID:                  /subscriptions/98836e77-374f-4e3b-8ea4-20ff30ca1218/resourceGroups/dv-k8-rg/providers/Microsoft.Compute/virtualMachines/k8s-agent-4EF87F69-0
ProviderID:                  azure:///subscriptions/98836e77-374f-4e3b-8ea4-20ff30ca1218/resourceGroups/dv-k8-rg/providers/Microsoft.Compute/virtualMachines/k8s-agent-4EF87F69-0
Non-terminated Pods:         (6 in total)
  Namespace                  Name                                    CPU Requests  CPU Limits  Memory Requests  Memory Limits
  ---------                  ----                                    ------------  ----------  ---------------  -------------
  kube-system                heapster-342135353-5rrth                138m (2%)     138m (2%)   294Mi (0%)       294Mi (0%)
  kube-system                kube-dns-v20-3003781527-3v8mp           110m (1%)     0 (0%)      120Mi (0%)       220Mi (0%)
  kube-system                kube-dns-v20-3003781527-9m4r1           110m (1%)     0 (0%)      120Mi (0%)       220Mi (0%)
  kube-system                kube-proxy-dr925                        100m (1%)     0 (0%)      0 (0%)           0 (0%)
  kube-system                kubernetes-dashboard-924040265-m8200    0 (0%)        0 (0%)      0 (0%)           0 (0%)
  kube-system                tiller-deploy-677436516-svlx8           50m (0%)      50m (0%)    150Mi (0%)       150Mi (0%)
Allocated resources:
  (Total limits may be over 100 percent, i.e., overcommitted.)
  CPU Requests  CPU Limits  Memory Requests  Memory Limits
  ------------  ----------  ---------------  -------------
  508m (8%)     188m (3%)   684Mi (1%)       884Mi (1%)
Events:
  Type    Reason                   Age                 From                              Message
  ----    ------                   ----                ----                              -------
  Normal  NodeHasSufficientDisk    37m (x21 over 54m)  kubelet, k8s-agent-4ef87f69-0     Node k8s-agent-4ef87f69-0 status is now: NodeHasSufficientDisk
  Normal  NodeHasSufficientMemory  37m (x21 over 54m)  kubelet, k8s-agent-4ef87f69-0     Node k8s-agent-4ef87f69-0 status is now: NodeHasSufficientMemory
  Normal  NodeHasNoDiskPressure    37m (x21 over 54m)  kubelet, k8s-agent-4ef87f69-0     Node k8s-agent-4ef87f69-0 status is now: NodeHasNoDiskPressure
  Normal  Starting                 37m                 kube-proxy, k8s-agent-4ef87f69-0  Starting kube-proxy.
  Normal  NodeReady                37m                 kubelet, k8s-agent-4ef87f69-0     Node k8s-agent-4ef87f69-0 status is now: NodeReady

Perhaps I need to install the GPU driver? I know that can be an issue with some of these. Or should I just go with AKS?