Creating GlusterFS Volumes¶
The volume module allows users to create volume using a specified list of hosts and bricks. Volume section supports the following variables:
volname - Name of the volume, if no name is provided gdeploy generates a volume name.
action - Action supports the following values create, delete, add-brick, remove-brick, rebalance, and set.
brick_dirs - This variable specifies the brick directories to use. The brick_dirs variable can take values in ip:brick_dir format or just brick_dir format. For example:
brick_dirs=10.0.0.1:/mnt/data1/1,10.0.0.2:/mnt/data2/2
Or
brick_dirs=/mnt/data1/1,/mnt/data2/2
transport - The transport type. Possible values are tcp,tcp,rdma,rdma
replica_count - The replication count for replica volumes.
force - If set to yes, force is used while creating volumes.
disperse - Identifies if the volume should be disperse. Possible options are [yes, no].
disperse_count - Optional argument. If none given, the number of bricks specified in the commandline is taken as the disperse_count value.
redundancy_count - If redundancy_count is not specified, and if disperse is yes, it’s default value is computed so that it generates an optimal configuration.
Example 1:
[volume]
action=create
volname=foo
transport=tcp
replica_count=2
force=yes
Example 2:
[backend-setup]
devices=sdb,sdc
vgs=vg1,vg2
pools=pool1,pool2
lvs=lv1,lv2
brick_dirs=/gluster/brick/brick{1,2}
# If backend-setup is different for each host
# [backend-setup:10.70.46.13]
# devices=sdb
# brick_dirs=/gluster/brick/brick1
#
# [backend-setup:10.70.46.17]
# devices=sda,sdb,sdc
# brick_dirs=/gluster/brick/brick{1,2,3}
#
[volume]
action=create
volname=sample_volname
replica=yes
replica_count=2
force=yes