glusterfs vagrant POC

3 nodos, cae 1
- los otros dos continúan funcionando
- al reincorporarse el tercero, restituye la información
2 nodos, añadir tercero posteriormente
- gluster peer probe node3
- gluster volume add-brick gv0 replica 3 node3:/exports/sdb1/brick1
- gluster volume status

1 nodo, añadir segundo (pruebas processing 3)

no deja crear un volumen con replica si no hay nodos para soportarlo
no deja crear un volumen con replica 1
arbiter mode con replica 2?? → NO

node 1:

sudo gluster volume create gv0 node1:/exports/sdb1/brick1 # volume create: gv0: success: please start the volume to access data
sudo gluster volume start gv0 # volume start: gv0: success
sudo mount -t glusterfs node1:/gv0 /mnt

node 2:

sudo gluster volume status
sudo mount -t glusterfs node2:/gv0 /mnt

node 1:

sudo gluster peer probe node2 #peer probe: success
sudo gluster volume add-brick gv0 replica 2 node2:/exports/sdb1/brick1 # volume add-brick: success

Replica 2 volumes are prone to split-brain. Use Arbiter or Replica 3 to avoid this. See: http://docs.gluster.org/en/latest/Administrator%20Guide/Split%20brain%20and%20ways%20to%20deal%20with%20it/.

require: vagrant plugin install vagrant-hosts

Vagrantfile

VAGRANTFILE_API_VERSION = "2"
 
cluster = {
  "node1" => { :ip => "10.0.0.10", :cpus => 1, :mem => 1024 },
  "node2" => { :ip => "10.0.0.20", :cpus => 1, :mem => 1024 },
  "node3" => { :ip => "10.0.0.30", :cpus => 1, :mem => 1024 }
}
 
Vagrant.configure(VAGRANTFILE_API_VERSION) do |config|
 
  cluster.each_with_index do |(hostname, info), index|
 
    config.vm.define hostname do |cfg|
 
      cfg.vm.provider :virtualbox do |vb, override|
        config.vm.box = "debian/buster64"
        override.vm.network :private_network, ip: "#{info[:ip]}"
        override.vm.hostname = hostname
        vb.name = hostname
        vb.customize ["modifyvm", :id, "--memory", info[:mem], "--cpus", info[:cpus], "--hwvirtex", "on"]
 
        file_to_disk = 'data/' + hostname + '-disk.vdi'
        unless File.exist?(file_to_disk)
          vb.customize ['createhd', 
                        '--filename', file_to_disk, 
                        '--size', 5 * 1024]
        end
        vb.customize ['storageattach', :id, 
                      '--storagectl', 'SATA Controller', 
                      '--port', 1, 
                      '--device', 0, 
                      '--type', 'hdd', 
                      '--medium', file_to_disk]
 
      end # end provider
 
      config.vm.provision 'shell', path: './data/allnodes.sh'
      config.vm.provision :hosts, :sync_hosts => true
 
      node_script = './data/' + hostname + '.sh'
      if File.exists?(node_script) then
        config.vm.provision "file", source: node_script, destination: "$HOME/"
      else
        config.vm.provision "shell", inline: "echo SCRIPT not found!"
      end
 
    end # end config
 
  end # end cluster
end

allnodes.sh

[[ ! -f /etc/provision_env_disk_added_date ]] && {
 
  echo 'type=83' | sudo sfdisk /dev/sdb
  sudo mkfs.ext4 /dev/sdb1
  sudo mkdir -p /exports/sdb1
  echo "/dev/sdb1      /exports/sdb1   ext4 user,auto,defaults     0       0" | sudo tee --append /etc/fstab
  sudo date > /etc/provision_env_disk_added_date
 
}
 
mount /exports/sdb1
sudo mkdir /exports/sdb1/brick1
 
sudo apt-get -y install gnupg
wget -O - https://download.gluster.org/pub/gluster/glusterfs/7/rsa.pub | sudo apt-key add -
# DEBID=$(grep 'VERSION_ID=' /etc/os-release | cut -d '=' -f 2 | tr -d '"')
# DEBVER=$(grep 'VERSION=' /etc/os-release | grep -Eo '[a-z]+')
# DEBARCH=$(dpkg --print-architecture)
# echo "deb https://download.gluster.org/pub/gluster/glusterfs/LATEST/Debian/${DEBID}/${DEBARCH}/apt ${DEBVER} main" | sudo tee /etc/apt/sources.list.d/gluster.list
echo "deb https://download.gluster.org/pub/gluster/glusterfs/LATEST/Debian/10/amd64/apt buster main" | sudo tee /etc/apt/sources.list.d/gluster.list
 
sudo apt-get -y update
 
sudo apt-get -y install glusterfs-server
sudo systemctl start glusterd
sudo systemctl enable glusterd

node1.sh

#!/bin/bash
 
sudo gluster peer probe node2
sudo gluster peer probe node3
 
sudo gluster volume create gv0 replica 3 \
  node1:/exports/sdb1/brick1 \
  node2:/exports/sdb1/brick1 \
  node3:/exports/sdb1/brick1
 
sudo gluster volume start gv0
 
sudo mount -t glusterfs node1:/gv0 /mnt

node{2,3}.sh

#! /bin/bash
 
MY_HOST=$(cat /etc/hostname)
sudo mount -t glusterfs ${MY_HOST}:/gv0 /mnt

en lugar de node1, podría ser cualquier otro de los dos nodos

glusterfs vagrant POC

tests

install

miguel angel torres egea