install oVirt 3.6 on Centos 7

I needed a quick set-up for some lab virtual machines so I used a laptop I had  and played with oVirt 3.6 over the past few days and things look promising. In then end after experimenting with Vmware ESX on the  laptop I settled with VirtualBox running in Centos7.

This is my walkthrough as I was installing oVirt 3.6, running everything on a single 1tb hard disk. They are my working notes using mainly

First install Centos7 minimal install, ensure that the network interface is on (off by default), you will need at least 10Gb in /var/tmp.

(optional) update to the latest available version:

# yum -y update

check the hostname:

hostnamectl status
hostnamectl set-hostname [new host name]

edit /etc/hosts file and add the entries for the host and engine:

# vi /etc/hosts
# systemctl restart systemd-hostnamed

disable NetworkManager:

# service NetworkManager stop
# chkconfig NetworkManager off
# service network start
# chkconfig network on

add correct gpg key for glusterfs:

# yum install -y wget
# wget
# rpm –import ./
# rm -f ./

install required apps (this will take a while):

# yum localinstall -y
# yum install -y ovirt-hosted-engine-setup screen glusterfs-server nfs-utils vdsm-gluster system-storage-manager mailx
# yum install -y ovirt-engine-appliance

create a partition for the glusterfs storage

# fdisk /dev/sda
# partprobe

list all partitions:

# lsblk

set-up gluster:

# pvcreate /dev/sda6
# ssm add -p vg_gluster /dev/sda6
# ssm create -p vg_gluster –fstype xfs -n lv_gluster

Add line to /etc/fstab:  /dev/mapper/vg_gluster-lv_gluster  /gluster  defaults  0 0

# mkidr /gluster
# mount /gluster
# mkdir -p /gluster/{data,engine,iso}/brick
# systemctl start glusterd && systemctl enable glusterd

Create the gluster “bricks”

# gluster volume create engine $HOSTNAME:/gluster/engine/brick
# gluster volume create data $HOSTNAME:/gluster/data/brick
# gluster volume create iso $HOSTNAME:/gluster/iso/brick
# gluster volume set engine group virt
# gluster volume set engine storage.owner-uid 36 && gluster volume set engine storage.owner-gid 36
# gluster volume set data group virt
# gluster volume set data storage.owner-uid 36 && gluster volume set data storage.owner-gid 36
# gluster volume set iso group virt
# gluster volume set iso storage.owner-uid 36 && gluster volume set iso storage.owner-gid 36

Start the Gluster volumes

# gluster volume start engine && gluster volume start data && gluster volume start iso

Check the Gluster volume status and availability

# gluster volume status
# showmount -e $HOSTNAME

deploy ovirt engine:

# hosted-engine –deploy


*** In case the engine deployment fails and you want to run it again, you can simply delete the host:/engine volume and recreate it.

# gluster volume stop engine
# gluster volume delete engine
# rm -rf /gluster/engine

and then recreate it as per above



Leave a Reply

Your email address will not be published. Required fields are marked *