Install GlusterFS from Source

From GlusterDocumentation

Jump to: navigation, search

1 Requirements
2 Download
- 2.1 Get GlusterFS
- 2.2 Get Patched Fuse (Completely Optional Step)
3 Install
- 3.1 FUSE
- 3.2 GlusterFS
4 Execution
- 4.1 Example
  - 4.1.1 Server machine: [192.168.0.1]
  - 4.1.2 Client machine: [192.168.0.2]
5 Something isnt working right
6 Refer

[edit]

Requirements

autotools (if you are compiling from tla)
libtool
gcc
flex
bison
byacc
kernerl-source for the $(uname -r) kernel (That means the currently running kernel, this is required if you need to compile fuse, otherwise, we don't need it).

[edit]

Download

[edit]

Get GlusterFS

Download GlusterFS from the following links

Or download via tla

$ tla register-archive http://arch.savannah.gnu.org/archives/gluster/
$ tla get -A gluster@sv.gnu.org glusterfs--mainline--2.5 glusterfs

[edit]

Get Patched Fuse (Completely Optional Step)

Download patched fuse from the following links

The patched version of fuse is well suited for use with GlusterFS, as:

It supports higher IO buffer size, which gives increased IO performance.
It provides flock() syscall, which provides you with advisory locks functionality. This is not present in regular fuse tarball.
Supports O_DIRECT flag handling in open() calls

Please note that if you intend to use the patched fuse module that you *MUST* make sure your kernel does not have one compiled in as this will override the loading of the updated module.

NOTE: Any new kernel (>2.6.25), please use custom fuse only.

[edit]

Install

NOTE: You need to be logged in as 'root' to run the following commands.

[edit]

FUSE

This section describes the installation procedure for 'fuse' from source tarball.

On the client machine, install the fuse package. (Make sure that '--prefix' option is set to the earlier fuse installation path).

NOTE: To compile fuse's kernel module you need to have the kernel source for your already running kernel.
NOTE: For kernel version greater than or equal to 2.6.25, the fuse kernel module won't compile. You need to use the module comes with kernel.
NOTE: It is advised to use kernel module and libfuse both from single fuse package. Otherwise, you may get some problems. (Not verified, but a better practice).

$ tar -xzf fuse-2.7.2glfs9.tar.gz
$ cd fuse-2.7.2glfs9
$ ./configure  --prefix=/usr  --enable-kernel-module
$ make install
$ ldconfig
$ depmod -a
$ rmmod fuse
$ modprobe fuse
$ echo modprobe fuse >> /etc/rc.modules
$ chmod +x /etc/rc.modules

GlusterFS works with fuse version 2.6.0 and above. But with custom fuse versions, there are some issues seen (Input/Output Errors). Hence it is better to use glusterfs patched fuse or FUSE version >= 2.7.2

[edit]

GlusterFS

Now, untar and install the glusterfs package.

$ tar -xzf glusterfs-1.3.10.tar.gz
$ cd glusterfs-1.3.10
$ ./configure  
$ make
# sudo make install

Congratulations :) You are done with 'glusterfs' installation.

[edit]

Execution

After installation is complete, the problem that the majority of people are faced with is how to get glusterfs working. To run GlusterFS, you need a volume specification file, or spec file, which defines the behavior and features for glusterfs.

We will start with a barebones spec file (this spec file is very basic and it is just to get the feel of GlusterFS).

NOTE: On many systems, modifying (or turning off) iptables may be required to get GlusterFS working.

$ glusterfs --help

[edit]

Example

In this basic example, let us assume you have two machines, '192.168.0.1' and '192.168.0.2'. Let 192.168.0.1 be the server and 192.168.2 be the client.

NOTE: you must change the IP address in the spec file according to your network configuration or else to test, you can use the system's localhost IP address (i.e. 127.0.0.1)

[edit]

Server machine: [192.168.0.1]

NOTE: After editing the file it should have the content as shown by the cat command, and this holds true for all the given examples below.

$ emacs /etc/glusterfs/glusterfs-server.vol
$ cat /etc/glusterfs/glusterfs-server.vol
volume brick
  type storage/posix
  option directory /tmp/export
end-volume

volume server
  type protocol/server
  subvolumes brick
  option transport-type tcp/server # For TCP/IP transport
  option auth.ip.brick.allow *
end-volume

$ glusterfsd -f /etc/glusterfs/glusterfs-server.vol

[edit]

Client machine: [192.168.0.2]

$ mkdir /mnt/glusterfs
$ emacs /etc/glusterfs/glusterfs-client.vol
$ cat /etc/glusterfs/glusterfs-client.vol
volume client
  type protocol/client
  option transport-type tcp/client
  option remote-host 192.168.0.1
  option remote-subvolume brick
end-volume

$ glusterfs -f /etc/glusterfs/glusterfs-client.vol /mnt/glusterfs

You should now be able to see the exported directory '192.168.0.1:/tmp/export' as /mnt/glusterfs on the client machine :O

[edit]