[DRBD-user] 4-way redundancy problem

Lars Ellenberg lars.ellenberg at linbit.com
Fri Jul 9 18:48:00 CEST 2010

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


On Fri, Jul 09, 2010 at 06:38:04PM +0200, Bart Coninckx wrote:
> On Wednesday 07 July 2010 18:23:53 Philippe Blondin wrote:
> > Hi,
> > 
> > I'm trying to setup a 4-way redundancy using stacked DRBD resources.
> > Unfortunately not much information is available, basically I'm trying to
> > fallow the documentation for creating a three-node setup (except that
> > the remote half of the DRBD configuration is also stacked, not just the
> > local cluster) as written here :
> > http://www.drbd.org/users-guide/s-pacemaker-stacked-resources.html
> > 
> > My setup is very similar to the fallowing figure except that all my
> > nodes are on the same Cluster (same site).
> > http://www.drbd.org/users-guide/drbd-resource-stacking-pacemaker-4nodes.png
> > 
> > Here's the interesting part of my DRBD.conf
> > 
> > ############################################
> > 
> > # Replication between Node1 <==> Node2
> > resource mysql-left {
> >   meta-disk internal;
> >   device /dev/drbd0;
> > 
> >   on node1 {
> >     disk      /dev/mapper/VolGroup00-data;
> >     address   192.168.2.90:7789;
> >   }
> >   on node2 {
> >     disk      /dev/mapper/VolGroup00-data;
> >     address   192.168.2.91:7789;
> >   }
> > }
> > 
> > # Replication between Node3 <==> Node4
> > resource mysql-right{
> >   meta-disk internal;
> >   device /dev/drbd0;
> > 
> >   on node3 {
> >     disk      /dev/mapper/VolGroup00-data;
> >     address   192.168.2.92:7789;
> >   }
> >   on node4 {
> >     disk      /dev/mapper/VolGroup00-data;
> >     address   192.168.2.93:7789;
> >   }
> > }
> > 
> > # Replication between stacked devices:
> > #                       Left  <=====>    Right
> > #         Stacked Node[1-2]               Stacked Node[3-4]
> > resource mysql-stacked{
> >   meta-disk internal;
> >   device /dev/drbd10;    # Using another device for stacked DRBD
> > 
> >   stacked-on-top-of mysql-left {
> >     address   10.0.0.10:7789;      # Using a Heartbeat IP, so it can run
> > either on Node1 or Node2
> >   }
> >   stacked-on-top-of mysql-right {
> >     address   10.0.0.20:7789;      # Using a Heartbeat IP, so it can run
> > either on Node3 or Node4
> >   }
> > }
> > 
> > ############################################
> > 
> > Here's the step I've done so far:
> > 
> > 1. I've prepare both left and right side (lower level) by running
> > respectively drbdadm create-md mysql-left on node1 and node2, and
> > drbdadm create-md mysql-right on node3 and node4.
> > 2. I've sync both sides.
> > 3. I've setup pacemaker so my Heartbeat IP is running on node1
> > (10.0.0.10) and node3 (10.0.0.20)
> > Online: [ node3 node2 node4 node1 ]
> > 
> > res_ClusterIP_stacked_right     (ocf::heartbeat:IPaddr2):       Started
> > node3
> > res_ClusterIP_stacked_left      (ocf::heartbeat:IPaddr2):       Started
> > node1
> > 
> > 3. Now I'm trying to prepare my stacked resource, so I put Node1 in
> > primary mode (drbdadm primary mysql-left), but after when I want to
> > initialize the internal data here's what happen ==>
> > 
> > [root at node1 etc]# drbdadm --stacked create-md mysql-stacked
> > You want me to create a v08 style flexible-size internal meta data block.
> > There appears to be a v08 flexible-size internal meta data block
> > already in place on /dev/drbd0 at byte offset 11374563328
> > Do you really want to overwrite the existing v08 meta-data?
> > [need to type 'yes' to confirm]
> > 
> > It tries to overwrite my internal data for my /dev/drbd0 device.. I'm
> >  pretty sure im missing something important here so I'm hoping someone can
> >  enlighten me? It's the first time I'm trying to setup stacked resource,
> >  please excuse my noobness ;P
> > 
> > Please don't hesitate to ask if you need any additionnal information, any
> >  help is very appreciated!
> > 
> > Thanks
> > 
> > -Philippe Blondin
> > 
> 
> I seem to have understood more than three is not supported. Or am I wrong?

Oh, you can stack on stack on stack...
If you go more than two levels, you will hit limitations of the conf
file syntax (but you can try and figure out how to do it using
drbdsetup directly).
And of course, it is more complex to set up, to handle and to
administrate than a two node cluster.
But, sure,
> > http://www.drbd.org/users-guide/s-pacemaker-stacked-resources.html
> > http://www.drbd.org/users-guide/drbd-resource-stacking-pacemaker-4nodes.png
is working just fine.

-- 
: Lars Ellenberg
: LINBIT | Your Way to High Availability
: DRBD/HA support and consulting http://www.linbit.com

DRBD® and LINBIT® are registered trademarks of LINBIT, Austria.
__
please don't Cc me, but send to list   --   I'm subscribed



More information about the drbd-user mailing list