White Papers

Dell HPC NFS Storage Solution - High Availability Configurations
Page 52
8) On the active server, create a volume group across all four LUNS. Use the same VG name as before;
else the /etc/cluster/cluster.conf file will need to be edited.
vgcreate DATA_VG /dev/mapper/mpath2 /dev/mapper/mpath3
/dev/mapper/mpath4 /dev/mapper/mpath5
9) On the active server, create a logical volume using this volume group.
lvcreate -i 4 -I 1024 -l 100%FREE DATA_VG -n DATA_LV
10) Set up HA LVM
 On the active server, edit /etc/lvm/lvm.conf and change the volume list to
volume_list = [ “VolGroup00” , “@active” ]
where VolGroup00 is the volume group that contains the “/” file system for the OS.
“active” is the name of the server as defined in the cluster.conf file.
 On the passive server, edit /etc/lvm/lvm.conf and change the volume list to
volume_list = [ “VolGroup00” , “@passive” ]
where VolGroup00 is the volume group that contains the “/” file system for the OS.
“passive” is the name of the server as defined in the cluster.conf file.
 On both servers, remake the initrd
mkinitrd -f /boot/initrd-2.6.18-194.el5.img 2.6.18-194.el5
11) On the active server, create the file system
mkfs.xfs -l size=128m /dev/DATA_VG/DATA_LV
12) Restart the cluster. This will activate the HA LVM, mount the XFS file system, export it via NFS and
create the floating resource IP.
Simultaneously on both servers:
service cman start
service rgmanager start
13) Check the new file system size. df –h should show the file system mounted and the total size of
the file system to be close to 80TB, double what it was earlier.
14) Restore user data.