Scaling OwnCloud with Red Hat Storage

1 . Replicated storage using Red Hat Storage Server

If you’re a regular reader of my articles you may recall that I have already documented Red Hat Storage Server in the past. In this article only minor changes have been made.

That being said, I’ll still cover the process.

Once you have your 3 Red Hat Storage Server nodes build you can do the following on all 3 nodes.

Note: I have a 100GB disk on each server appearing as /dev/vdb.

Create local volume

It is entirely up to you how you chose to prepare your underlying storage, however in this example, I am using LVM on top of my block devices.

[root@rhs01 ~]# pvcreate /dev/vdb
Writing physical volume data to disk "/dev/vdb"
Physical volume "/dev/vdb" successfully created
[root@rhs01 ~]# vgcreate vg_rhs /dev/vdb
Volume group "vg_rhs" successfully created
[root@rhs01 ~]# lvcreate -L 100G /dev/vg_rhs --name lv_rhsvol1
Logical volume "lv_rhsvol1" created

Format local volume

[root@rhs01 ~]# mkfs.xfs -i size=512 /dev/vg_rhs/lv_rhsvol1
meta-data=/dev/vg_rhs/lv_rhsvol1  isize=512    agcount=4, agsize=3276544 blks
=                       sectsz=512   attr=2
data     =                       bsize=4096   blocks=13106176, imaxpct=25
=                       sunit=0      swidth=0 blks
naming   =version 2              bsize=4096   ascii-ci=0
log      =internal log           bsize=4096   blocks=6399, version=2
=                       sectsz=512   sunit=0 blks, lazy-count=1
realtime =none                   extsz=4096   blocks=0, rtextents=0

Mount local volume

[root@rhs01 ~]# mkdir /rhs/vol1
[root@rhs01 ~]# echo "/dev/vg_rhs/lv_rhsvol1   /rhs/vol1                 xfs     defaults,allocsize=4096 0 0" >> /etc/fstab
[root@rhs01 ~]# mount -a
[root@rhs01 ~]# df -mhP
Filesystem                     Size Used Avail Use% Mounted on
/dev/mapper/vg_base-lv_root    8.7G 1.9G  6.4G  23%    /
tmpfs                          499M    0  499M   0%    /dev/shm
/dev/vda1                      194M  51M  134M  28%    /boot
/dev/mapper/vg_rhs-lv_rhsvol1  100G  33M  100G   1%    /rhs/vol1
[root@rhs01 ~]#

Lastly, create a local folder inside of your new volumes to be used for our web content. Here I am creating “www” inside of “/rhs/vol1”

[root@rhs01 ~]# mkdir /rhs/vol1/www

You should have your local storage created, formatted and presented on all your Red Hat Storage Server nodes, now its time to create them into a single “Gluster” volume. In this example, we will be created a “replicated” volume as opposed to a “distributed” volume.

 

Note: From only one of the Red Hat Storage Server nodes, run the following.

Probe other servers

You will need to probe the other servers prior to creating a volume.

gluster peer probe rhs02.example.com
gluster peer probe rhs03.example.com

Once you have probed all the servers you intend to use, you can list them with the following.
Be aware that if you see any IP addresses instead of hostnames, you have not configured your DNS resolution for those hosts correctly.

[root@rhs01 ~]# gluster peer status
Number of Peers: 2<
Hostname: rhs02.example.com
Uuid: 26f2c32e-63b2-4d31-a6f7-1819bb849e74
State: Peer in Cluster (Connected)
Hostname: rhs03.example.com
Uuid: 3b3fefb5-561d-4da3-85eb-52f2bb679c86
State: Peer in Cluster (Connected)
[root@rhs01 ~]#

Create Gluster Volume

Now lets bring all the above together to form a replicated Gluster volume.

To do this, run the following, notice that we are forming a volume with “3” replica’s as well as pointing it to 3 server names.

[root@rhs01 ~]# gluster volume create WWW replica 3 transport tcp rhs01.example.com:/rhs/vol1/www rhs02.example.com:/rhs/vol1/www rhs03.example.com:/rhs/vol1/www
volume create: WWW: success: please start the volume to access data

Start Gluster Volume

The last step here is to start the volume, although the volume is created, it is not accessible for use until it is started.

[root@rhs01 ~]# gluster volume start WWW
volume start: WWW: success

Once you have started the volume, you can verify its details with the below.

[root@rhs01 ~]# gluster volume info WWW
Volume Name: WWW
Type: Replicate
Volume ID: 36b3f37e-19fb-4d97-99e3-f287f70d098d
Status: Started
Number of Bricks: 1 x 3 = 3
Transport-type: tcp
Bricks:
Brick1: rhs01.example.com:/rhs/vol1/www
Brick2: rhs02.example.com:/rhs/vol1/www
Brick3: rhs03.example.com:/rhs/vol1/www
[root@rhs01 ~]#