I have two nodes (running Armbian Bionic, based on Ubuntu 18.04) set up to create a distributed Gluster storage, provisioned via Ansible.
I think the relevant part of the Ansible playbook is this:
- name: Configure gluster volume
gluster_volume:
state: present
name: "{{ gluster_volume_name }}"
brick: "{{ gluster_brick_dir }}"
cluster: "{{ groups.glustergroup | join(',') }}"
host: "{{ inventory_hostname }}"
force: yes
run_once: true
Where glustergroup is the list of the two nodes in my Ansible inventory file.
On both nodes, when I run gluster volume info I see that the volume was created successfully, and it can be mounted via mount -t glusterfs. Here's the output of gluster volume info:
Volume Name: [my volume name]
Type: Distribute
Volume ID: edbc9b23-6252-4725-9652-e46c280dae2b
Status: Started
Snapshot Count: 0
Number of Bricks: 2
Transport-type: tcp
Bricks:
Brick1: [node 1]:/bricks/brick0
Brick2: [node 2]:/bricks/brick0
Options Reconfigured:
transport.address-family: inet
nfs.disable: on
I noticed that nfs.disable: on, but I want my clients to mount the volume via NFS. So, I ran gluster volume set [my volume name] nfs.disable off, then rebooted the nodes for good measure.
Now, I can see this from gluster volume status:
Status of volume: [my volume name]
Gluster process TCP Port RDMA Port Online Pid
------------------------------------------------------------------------------
Brick [node 1]:/bricks/brick0 49152 0 Y 2338
Brick [node 2]:/bricks/brick0 49152 0 Y 1526
NFS Server on localhost N/A N/A N N/A
NFS Server on [node 2] N/A N/A N N/A
Notice that the NFS Servers are still not online in either node.
What else do I need to do to turn on NFS support, so that clients can mount this volume via /etc/fstab? Is there anything I can/should modify in the Ansible playbook to enable this? Thank you.
