Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Michael, I have a similar setup, however, I am going with a simpler configuration. I have two Xen hypervisor machines, each with a 1TB volume. I use LVM on top of the 1TB volume to carve out LVs for each VM harddrive. Then I have DRBD replicating each LV. So, currently I have 14 DRBD devices, I add a new DRBD resource whenever I create a new VM. This allows each VM to migrate from one hypervisor to the other independently. All the DRBD resources are setup for dual primary, this is needed to support Xen live migration. I let Hearbeat manage the VMs and I use the drbd: storage type for the Xen VMs, so Xen can handle the DRBD resources. This gives me failover for all the VMs, as well as manual live migration. Currently I run half the VMs on each hypervisor 7/7, to spread the load, of course Hearbeat will boot up the VMs on the remaining hypervisor if one of the systems fail. When I perform maintenance, I can put a Heartbeat node into standby and the VMs live migrate. This has been a very stable configuration for me. On Mon, May 10, 2010 at 11:30 AM, Michael Iverson <dr.michael.iverson at gmail.com> wrote: > I'm just about to embark on my first adventure with DRBD, and I'm > interested in some advice on the most reliable approach. > > I'm going to have two servers configured as a DRBD cluster, with > either a gigabit ethernet or infiniband link between them. This > cluster will host a number of virtual machine images, using KVM, > likely using Convirt as a management tool. I will have one or more > machines in addition to the two main machines, which will be used to > host images during maintenance or in the event of a machine failure. > > My question is how best to export file system images to the various > machines. My first thought was to use a primary/primary setup and > export LVM volumes via multipath iSCSI. The advantage here is that the > load on the network and file system is spread evenly, the > configuration allows for failover from just about any conceivable > failure, and direct LVM access minimizes the file system overhead. The > configuration seems to be fairly complex, but I've tackled similar > complexity in the past without concern. > > However, In my research, I've come across articles like this: > http://87.230.77.133/blog/?p=6. This is causing me to second guess as > to whether this is the best approach to take. I'm struggling to find > consensus on the web as to the best path to take. > Obviously, I have other options at my disposal, including: > > 1. Add OCFS2 to the mix, and use file-based images instead of LVM volumes. > 2. Abandon iSCSI for NFS. > 3. Abandon multipath in favor of a simple heartbeat failover. > 4. Switch to a primary/secondary configuration. > > Does anyone have opinions on the most appropriate path to take? > > -- > Dr. Michael Iverson > Director of Information Technology > Hatteras Printing > _______________________________________________ > drbd-user mailing list > drbd-user at lists.linbit.com > http://lists.linbit.com/mailman/listinfo/drbd-user >