On Sat, 15 Mar 2008, M. David Peterson wrote:
So I have been able to successfully get heartbeat up and running
between two EC2 instances in which DRBD keeps two ephemeral drives
in sync with one another. /dev/drbd0 is mounted @ /data, and
using unicast the system has successfully remounted /dev/drbd0 on
system2 when system1 goes down, removing on system2 and remounting
on system1 when it comes back up.
However, I haven't quite yet figured out how to handle the virtual
IP allocation so at the moment there is no way to mount the drive
via NFS using anything other than the "static" IP of server1.
I've been playing around with OpenVPN, attempting to use the tun0
device instead of eth0 with IPAddr2, but haven't had any luck.
You should be able to add an IPaddr or IPaddr2 resource to assign a
floating IP to the master drbd server. Add colocation and order
constraints to make sure that the drbd resource comes up first.
The general approach is all on the page below; adapt as needed for
your resources.
http://www.linux-ha.org/DRBD/HowTov2
-- Michael
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
