This post has NOT been accepted by the mailing list yet.
So, I've been struggling for about 2 weeks to cobble together the bits and bobs required to create a highly available iSCSI Target cluster in AWS. I have a Pacemaker/Corosync cluster in place using DRBD for block-level replication of the EBS volumes used as target storage between the nodes. While I have managed to stitch together 4-5 how-tos to get to this point, I find myself struggling (conceptualizing and implementing) with how to make the final bits function in a multi-AZ implementation. Most guides have you set up instances in the same availability zone and the same subnet. This makes it easy to create a secondary "vip" address that the nodes can share for the ocf:heartbeat:IPaddr2 resource. This is, for obvious reasons, not the case when you're talking multi-AZ because IP subnets do not span availability zones. Can anyone walk me through this or point me somewhere that will? Thanks!