Installing and Configuring NVMe-oF Targets

Traditionally, block-level access to a Ceph storage cluster has been limited to (1) QEMU and librbd (which is a key enabler for adoption within OpenStack environments), and (2) the Linux kernel client. Starting with the Ceph Reef release, block-level access has been expanded to offer standard NVMe/TCP support, allowing wider platform usage and potentially opening new use cases.

Prerequisites

  • Red Hat Enterprise Linux/CentOS 8.0 (or newer); Linux kernel v4.16 (or newer)

  • A working Ceph Reef or later storage cluster, deployed with cephadm

  • NVMe-oF gateways, which can either be colocated with OSD nodes or on dedicated nodes

  • Separate network subnets for NVME-oF front-end traffic and Ceph back-end traffic

Explanation

The Ceph NVMe-oF gateway is both an NVMe-oF target and a Ceph client. Think of it as a “translator” between Ceph’s RBD interface and the NVME-oF protocol. The Ceph NVMe-oF gateway can run on a standalone node or be colocated with other daemons, for example on an OSD node. When colocating the Ceph NVMe-oF gateway with other daemons, ensure that sufficient CPU and memory are available. The steps below explain how to install and configure the Ceph NVMe/TCP gateway for basic operation.

Installation

Complete the following steps to install the Ceph NVME-oF gateway:

  1. Create a pool in which the gateways configuration can be managed:

    ceph osd pool create NVME-OF_POOL_NAME

  2. Enable RBD on the NVMe-oF pool:

    rbd pool init NVME-OF_POOL_NAME

  3. Deploy the NVMe-oF gateway daemons on a specific set of nodes:

    ceph orch apply nvmeof NVME-OF_POOL_NAME --placement="host01, host02"

Configuration

Download the nvmeof-cli container before first use. To download it use the following command:

podman pull quay.io/ceph/nvmeof-cli:latest

  1. Create an NVMe subsystem:

    podman run -it quay.io/ceph/nvmeof-cli:latest --server-address GATEWAY_IP --server-port GATEWAY_PORT 5500 subsystem add --subsystem SUSYSTEM_NQN

    The subsystem NQN is a user defined string, for example nqn.2016-06.io.spdk:cnode1.

  2. Define the IP port on the gateway that will process the NVME/TCP commands and I/O:

    1. On the install node, get the NVME-oF Gateway name:

      ceph orch ps | grep nvme

    2. Define the IP port for the gateway:

      podman run -it quay.io/ceph/nvmeof-cli:latest --server-address GATEWAY_IP --server-port GATEWAY_PORT 5500 listener add --subsystem SUBSYSTEM_NQN --gateway-name GATEWAY_NAME --traddr GATEWAY_IP --trsvcid 4420

  3. Get the host NQN (NVME Qualified Name) for each host:

    cat /etc/nvme/hostnqn

    esxcli nvme info get

  4. Allow the initiator host to connect to the newly-created NVMe subsystem:

    podman run -it quay.io/ceph/nvmeof-cli:latest --server-address GATEWAY_IP --server-port GATEWAY_PORT 5500 host add --subsystem SUBSYSTEM_NQN --host "HOST_NQN1, HOST_NQN2"

  5. List all subsystems configured in the gateway:

    podman run -it quay.io/ceph/nvmeof-cli:latest --server-address GATEWAY_IP --server-port GATEWAY_PORT 5500 subsystem list

  6. Create a new NVMe namespace:

    podman run -it quay.io/ceph/nvmeof-cli:latest --server-address GATEWAY_IP --server-port GATEWAY_PORT 5500 namespace add --subsystem SUBSYSTEM_NQN --rbd-pool POOL_NAME --rbd-image IMAGE_NAME

  7. List all namespaces in the subsystem:

    podman run -it quay.io/ceph/nvmeof-cli:latest --server-address GATEWAY_IP --server-port GATEWAY_PORT 5500 namespace list --subsystem SUBSYSTEM_NQN