I've finally built my first cluster on rhel5 using gfs2 and drbd8 :-D
drbd makes a network raid1 between two physical distinguished block devices (eg. internal disks of two servers) and from the release 8.0 it supports the active/active configuration.
gfs2 permits to the two server machines a concurrent mount of the network replicated drbd device.
rhel5 is the OS I've installed on the two machines.
Practically with this stuff the servers share the same "storage". Here is a sample config file for drbd:
common { syncer { rate 100M; } }
resource r0 {
protocol C;
disk { on-io-error pass_on; }
net {
allow-two-primaries;
after-sb-0pri discard-least-changes;
after-sb-1pri discard-secondary;
after-sb-2pri violently-as0p;
}
on bsvm1.babel.int {
address 192.168.1.1:7790;
disk /dev/mapper/ddf1_ld0p5;
device /dev/drbd0;
meta-disk "internal";
}
on bsvm1.babel.int {
address 192.168.1.2:7790;
disk /dev/mapper/ddf1_ld0p5;
device /dev/drbd0;
meta-disk "internal";
}
}
Want to give it a try? Configure your cluster.xml file, start the drbd and cman services, create the filesystem on the drbd device and ... have fun!!