This tutorial shows how to do data striping (segmentation of logically sequential data, such as a single file, so that segments can be assigned to multiple physical devices in a round-robin fashion and thus written concurrently) across four single storage servers (running Debian Lenny) with GlusterFS. The client system (Debian Lenny as well) will be able to access the storage as if it was a local filesystem. GlusterFS is a clustered file-system capable of scaling to several peta-bytes. It aggregates various storage bricks over Infiniband RDMA or TCP/IP interconnect into one large parallel network file system. Storage bricks can be made of any commodity hardware such as x86-64 servers with SATA-II RAID and Infiniband HBA.
Please note that this kind of storage doesn’t provide any high-availability/fault tolerance features, as would be the case with replicated storage.
I do not issue any guarantee that this will work for you!
1 Preliminary Note
In this tutorial I use five systems, four servers and a client:
- server1.example.com: IP address 192.168.0.100 (server)
- server2.example.com: IP address 192.168.0.101 (server)
- server3.example.com: IP address 192.168.0.102 (server)
- server4.example.com: IP address 192.168.0.103 (server)
- client1.example.com: IP address 192.168.0.104 (client)
All five systems should be able to resolve the other systems’ hostnames. If this cannot be done through DNS, you should edit the /etc/hosts file so that it looks as follows on all five systems:
vi /etc/hosts
127.0.0.1 localhost.localdomain localhost 192.168.0.100 server1.example.com server1 192.168.0.101 server2.example.com server2 192.168.0.102 server3.example.com server3 192.168.0.103 server4.example.com server4 192.168.0.104 client1.example.com client1 # The following lines are desirable for IPv6 capable hosts ::1 localhost ip6-localhost ip6-loopback fe00::0 ip6-localnet ff00::0 ip6-mcastprefix ff02::1 ip6-allnodes ff02::2 ip6-allrouters ff02::3 ip6-allhosts |
(It is also possible to use IP addresses instead of hostnames in the following setup. If you prefer to use IP addresses, you don’t have to care about whether the hostnames can be resolved or not.)
2 Setting Up The GlusterFS Servers
server1.example.com/server2.example.com/server3.example.com/server4.example.com:
GlusterFS isn’t available as a Debian package for Debian Lenny, therefore we have to build it ourselves. First we install the prerequisites:
aptitude install sshfs build-essential flex bison byacc libdb4.6 libdb4.6-dev
Then we download the latest GlusterFS release from http://www.gluster.org/download.php and build it as follows:
cd /tmp
wget http://ftp.gluster.com/pub/gluster/glusterfs/2.0/LATEST/glusterfs-2.0.1.tar.gz
tar xvfz glusterfs-2.0.1.tar.gz
cd glusterfs-2.0.1
./configure –prefix=/usr > /dev/null
server1:/tmp/glusterfs-2.0.1# ./configure –prefix=/usr > /dev/null
GlusterFS configure summary
===========================
FUSE client : no
Infiniband verbs : no
epoll IO multiplex : yes
Berkeley-DB : yes
libglusterfsclient : yes
mod_glusterfs : no ()
argp-standalone : no
server1:/tmp/glusterfs-2.0.1#
make && make install
ldconfig
The command
glusterfs –version
should now show the GlusterFS version that you’ve just compiled (2.0.1 in this case):
server1:/tmp/glusterfs-2.0.1# glusterfs –version
glusterfs 2.0.1 built on May 29 2009 17:23:10
Repository revision: 5c1d9108c1529a1155963cb1911f8870a674ab5b
Copyright (c) 2006-2009 Z RESEARCH Inc. <http://www.zresearch.com>
GlusterFS comes with ABSOLUTELY NO WARRANTY.
You may redistribute copies of GlusterFS under the terms of the GNU General Public License.
server1:/tmp/glusterfs-2.0.1#
Next we create a few directories:
mkdir /data/
mkdir /data/export
mkdir /data/export-ns
mkdir /etc/glusterfs
Now we create the GlusterFS server configuration file /etc/glusterfs/glusterfsd.vol which defines which directory will be exported (/data/export) and what client is allowed to connect (192.168.0.104 = client1.example.com):
vi /etc/glusterfs/glusterfsd.vol
volume posix type storage/posix option directory /data/export end-volume volume locks type features/locks subvolumes posix end-volume volume brick type performance/io-threads option thread-count 8 subvolumes locks end-volume volume server type protocol/server option transport-type tcp/server option auth.addr.brick.allow 192.168.0.104 subvolumes brick end-volume |
Please note that it is possible to use wildcards for the IP addresses (like 192.168.*) and that you can specify multiple IP addresses separated by comma (e.g. 192.168.0.104,192.168.0.105).
Afterwards we create the system startup links for the glusterfsd init script…
update-rc.d glusterfsd defaults
… and start glusterfsd:
/etc/init.d/glusterfsd start
3 Setting Up The GlusterFS Client
client1.example.com:
On the client, we need to install fuse and GlusterFS. Instead of installing the libfuse2 package from the Debian repository, we install a patched version with better support for GlusterFS.
First we install the prerequisites again:
aptitude install sshfs build-essential flex bison byacc libdb4.6 libdb4.6-dev
Then we build fuse as follows (you can find the latest patched fuse version on ftp://ftp.zresearch.com/pub/gluster/glusterfs/fuse/):
cd /tmp
wget ftp://ftp.zresearch.com/pub/gluster/glusterfs/fuse/fuse-2.7.4glfs11.tar.gz
tar -zxvf fuse-2.7.4glfs11.tar.gz
cd fuse-2.7.4glfs11
./configure
make && make install
Afterwards we build GlusterFS (just like on the server)…
cd /tmp
wget http://ftp.gluster.com/pub/gluster/glusterfs/2.0/LATEST/glusterfs-2.0.1.tar.gz
tar xvfz glusterfs-2.0.1.tar.gz
cd glusterfs-2.0.1
./configure –prefix=/usr > /dev/null
make && make install
ldconfig
glusterfs –version
… and create the following two directories:
mkdir /mnt/glusterfs
mkdir /etc/glusterfs
Next we create the file /etc/glusterfs/glusterfs.vol:
vi /etc/glusterfs/glusterfs.vol
volume remote1 type protocol/client option transport-type tcp/client option remote-host server1.example.com option remote-subvolume brick end-volume volume remote2 type protocol/client option transport-type tcp/client option remote-host server2.example.com option remote-subvolume brick end-volume volume remote3 type protocol/client option transport-type tcp/client option remote-host server3.example.com option remote-subvolume brick end-volume volume remote4 type protocol/client option transport-type tcp/client option remote-host server4.example.com option remote-subvolume brick end-volume volume stripe type cluster/stripe option block-size 1MB subvolumes remote1 remote2 remote3 remote4 end-volume volume writebehind type performance/write-behind option window-size 1MB subvolumes stripe end-volume volume cache type performance/io-cache option cache-size 512MB subvolumes writebehind end-volume |
Make sure you use the correct server hostnames or IP addresses in the option remote-host lines!
That’s it! Now we can mount the GlusterFS filesystem to /mnt/glusterfs with one of the following two commands:
glusterfs -f /etc/glusterfs/glusterfs.vol /mnt/glusterfs
or
mount -t glusterfs /etc/glusterfs/glusterfs.vol /mnt/glusterfs
You should now see the new share in the outputs of…
mount
client1:~# mount
/dev/sda1 on / type ext3 (rw,errors=remount-ro)
tmpfs on /lib/init/rw type tmpfs (rw,nosuid,mode=0755)
proc on /proc type proc (rw,noexec,nosuid,nodev)
sysfs on /sys type sysfs (rw,noexec,nosuid,nodev)
udev on /dev type tmpfs (rw,mode=0755)
tmpfs on /dev/shm type tmpfs (rw,nosuid,nodev)
devpts on /dev/pts type devpts (rw,noexec,nosuid,gid=5,mode=620)
fusectl on /sys/fs/fuse/connections type fusectl (rw)
/etc/glusterfs/glusterfs.vol on /mnt/glusterfs type fuse.glusterfs (rw,max_read=131072,allow_other,default_permissions)
client1:~#
… and…
df -h
client1:~# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda1 29G 896M 27G 4% /
tmpfs 126M 0 126M 0% /lib/init/rw
udev 10M 80K 10M 1% /dev
tmpfs 126M 0 126M 0% /dev/shm
/etc/glusterfs/glusterfs.vol
105G 3.4G 96G 4% /mnt/glusterfs
client1:~#
(server1.example.com, server2.example.com, server3.example.com, and server4.example.com each have about 26GB of space for the GlusterFS filesystem, so that the resulting share has a size of about 4 x 26GB (105GB).)
Instead of mounting the GlusterFS share manually on the client, you could modify /etc/fstab so that the share gets mounted automatically when the client boots.
Open /etc/fstab and append the following line:
vi /etc/fstab
[...] /etc/glusterfs/glusterfs.vol /mnt/glusterfs glusterfs defaults 0 0 |
To test if your modified /etc/fstab is working, reboot the client:
reboot
After the reboot, you should find the share in the outputs of…
df -h
… and…
mount
4 Testing
Now let’s create a big test file on the GlusterFS share:
client1.example.com:
dd if=/dev/zero of=/mnt/glusterfs/test.img bs=1024k count=1000
ls -l /mnt/glusterfs
client1:~# ls -l /mnt/glusterfs
total 1028032
-rw-r–r– 1 root root 1048576000 2009-06-03 20:51 test.img
client1:~#
Now let’s check the /data/export directory on server1.example.com, server2.example.com, server3.example.com, and server4.example.com. You should see the test.img file on each node, but with different sizes (due to data striping):
server1.example.com:
ls -l /data/export
server1:~# ls -l /data/export
total 257008
-rw-r–r– 1 root root 1045430272 2009-06-03 20:51 test.img
server1:~#
server2.example.com:
ls -l /data/export
server2:~# ls -l /data/export
total 257008
-rw-r–r– 1 root root 1046478848 2009-06-03 20:55 test.img
server2:~#
server3.example.com:
ls -l /data/export
server3:~# ls -l /data/export
total 257008
-rw-r–r– 1 root root 1047527424 2009-06-03 20:54 test.img
server3:~#
server4.example.com:
ls -l /data/export
server4:~# ls -l /data/export
total 257008
-rw-r–r– 1 root root 1048576000 2009-06-03 20:02 test.img
server4:~#
5 Links
- GlusterFS: http://www.gluster.org/
- Debian: http://www.debian.org/