Pular para o conteúdo principal

NFS Volumes

Volumes provide persistent NFS storage hosted on a node. They can be mounted by compute instances (Docker Swarm or Kubernetes) across the cluster.

Create a volume​

nimbus volume create --name mydata --node NODE_ID --folder /exports/mydata

This sets up the NFS export on the node and makes it available for attachment.

List volumes​

nimbus volume list

Attach to a compute instance​

At creation time:

nimbus compute run --name app --swarm SWARM_ID \
  --image myapp --type small --volume VOL_ID:/data

To a running instance:

nimbus compute volume attach --instance INSTANCE_ID --volume VOL_ID:/data
nimbus compute volume detach --instance INSTANCE_ID --volume VOL_ID

How Docker volumes are mounted​

On Docker Swarm, each node gets a Docker volume for every active NFS volume. When a node is the same machine that hosts the NFS export, Nimbus creates a local bind-mount volume (type=none, o=bind) pointing directly to the export folder instead of mounting over NFS. This avoids loopback NFS traffic and gives better I/O performance for workloads that land on the storage node.

For all other nodes the volume is mounted over NFS as normal.

Attach to a Kubernetes cluster​

Creates a PersistentVolume and PersistentVolumeClaim backed by the NFS export:

nimbus k8s volume attach --cluster CLUSTER_ID --name VOL_ID --size 1Gi
nimbus k8s volume detach --cluster CLUSTER_ID --name VOL_ID

When a new Kubernetes cluster is created, all existing active volumes are automatically provisioned as PV/PVC on that cluster — no manual attach step needed.

Delete a volume​

Volumes must be detached from all resources before deletion:

nimbus volume delete VOL_ID