node-disk-manager (NDM) aims to make it easy to manage the disks attached to the node. It treats disks as resources that need to be monitored and managed just like other resources like CPU, Memory and Network. It contains a daemon which runs on each node, detects attached disks and loads them as BlockDevice objects (custom resource) into Kubernetes.
While PVs are well suited for stateful workloads, the BlockDevice objects are aimed towards helping hyper-converged Storage Operators by providing abilities like:
NDM has 2 main components:
and 2 optional components:
The design of the project is covered under this design proposal
The feedback of users / organizations currently using OpenEBS can be found here
Currently, the NDM project is in beta.
A detailed usage documentation is maintained in the wiki.
K8S : 1.18+
namespace, serviceAccount, configure filters in the node-disk-manager-config-map.kubectl create -f ndm-operator.yaml.kubectl to fetch BlockDevice Informationkubectl get blockdevices displays the blockdevices across the cluster, with NODENAME showing the node to which disk is attached, CLAIMSTATE showing whether the device is currently in use and STATE showing whether the device is connected to the node.kubectl get blockdevices -o wide displays the blockdevice along with the path at which the device is attached on the node.kubectl get blockdevices <blockdevice-cr-name> -o yaml displays all the details of the disk captured by ndm for given disk resource.Before building the image locally, you need to setup your development environment. The detailed instructions for setting up development environment, building and testing are available here.
OpenEBS welcomes your feedback and contributions in any form possible.