| saman_artorious |
07-04-2012 09:43 AM |
Glusterfs: how to make two systems share a folder?
I am trying to connect two system (let's say for time being) together such that it supports clustering.
for this I got the following packages:
Code:
# sudo apt-get install pacemaker sysv-rc-conf glusterfs-server glusterfs-examples glusterfs-client chkconfig nmap ntp
Next, I did the following for both system:
Code:
node[x]:~# mkfs.ext3 /dev/sd??
node[x]:~# blkid -g
node[x]:~# blkid /dev/sd?? >> /etc/fstab
node[x]:~# vi /etc/fstab
you must have a line like that :
UUID=9dc20d6c-a3d7-4667-a9b1-e8939a0473f1 /export ext3 defaults 0 2
node[x]:~# mount /export/
node[x]:~# mkdir /export/part1
Then, I added two files inside /etc/glusterfs. These are glusterd.vol & raid1.vol
Here's the contents of the two files, for the second system ip addresses may change though.
this is for glusterd.vol
Code:
volume management
type mgmt/glusterd
option working-directory /etc/glusterd
option transport-type socket,rdma
option transport.socket.keepalive-time 10
option transport.socket.keepalive-interval 2
end-volume
Quote:
the above lines are defaults, i mean, they're the only lines which are included in this file when you installed glusterfs package, I commented them first as I thought they wouldn't be needed, but then I got the endpoint connection error when tried to mount /mnt/glusterfs.
|
volume posix
type storage/posix
option directory /export/part1
end-volume
volume brick
type features/locks
subvolumes posix
end-volume
volume server
type protocol/server
subvolumes brick
option transport-type tcp
option transport.socket.bind-address 192.168.3.13 # system IP
option transport.socket.listen-port 820
option transport.socket.nodelay on
option auth.addr.brick.allow 127.0.0.1,192.168.3.13,192.168.3.99 #Systems IPs
end-volume
and this is the raid1.vol file
Code:
volume VolNode1
type protocol/client
option remote-subvolume brick
option transport-type tcp/client
option remote-host 192.168.3.13 # system 1 IP
option remote-port 820
option transport.socket.nodelay on
end-volume
volume VolNode2
type protocol/client
option remote-subvolume brick
option transport-type tcp/client
option remote-host 192.168.3.99 # system 2 IP
option remote-port 820
option transport.socket.nodelay on
end-volume
volume afr
type cluster/replicate
subvolumes VolNode2 VolNode1
option read-subvolume VolNode2
end-volume
volume wb
type performance/write-behind
subvolumes afr
option cache-size 4MB
end-volume
volume cache
type performance/io-cache
subvolumes wb
option cache-size 1024MB
option cache-timeout 60
end-volume
could anyone tell me how to make the two system work fine n share the folder? |
