[PVE-User] DRBD issue with prox.

[Robert Fantini]

On Mon, May 9, 2011 at 2:51 PM, Paul Dickson <assplove at gmail.com> wrote:
> This is more drbd related, but since I'm trying to do it on prox, i thought
> i'd try here.
>
> I'm trying to try out some of the features, including disk replication w/
> live migration, of proxmox.  Because of the way prox takes all drive space
> by default on install, i'm trying to work around that.  I've shrunk the
> default data LV which was taking up the entire VG.  I then created a second
> LV "/dev/pve/data2".  That LV is the base for the DRBD system.
>
> Relivant sections of the drbd.conf:
>         on proxmox1 {
>                 device /dev/drbd0;
>                 disk /dev/pve/data2;
>                 address 192.168.10.30:7788;
>                 meta-disk internal;
>         }
>         on proxmox2 {
>                 device /dev/drbd0;
>                 disk /dev/pve/data2;
>                 address 192.168.10.31:7788;
>                 meta-disk internal;
>         }
>
>
> So..
> PVName : /dev/sda2 -> VG name: pve -> LV Name: /dev/pve/data2 -> DRBD vol:
> /dev/drbd0 -> PV name dm-3  -> VG name: repl -> VM VOLUMES.
>
> The problem is when I create the "repl" vg, it isn't replicated.  I see DRBD
> replicating data and fully synced, however the VG is not replicated.
>
>
> Obviously I can start over and partition manually for proper DRBD
> configuration, but I'd prefer not to spend the time just for a proof of
> concept test.  I'm also curious if I am doing something to make the
> replication not work or if it just won't work using a DRBD device as the PV.
>
>
>
> _______________________________________________
> pve-user mailing list
> pve-user at pve.proxmox.com
> http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
>
>

I may not be able to help with why your setup is not working.   I'm
not a drbd dev, just a sysadmin who has used drbd + proxmox for over a
year.

 We use drbd + heartbeat for our data and proxmox vm's .

 our drdb is on a partition not a lvm.  the partition has an ext3 filesystem.

 we have high availability in that if the primary server breaks, the
secondary server starts vz  , qemu-server and gets i/p addresses.

 Are you using heartbeat along with drbd?