[DRBD-user] New storage: avoid split brains

Lars Ellenberg lars.ellenberg at linbit.com
Tue Oct 18 20:20:21 CEST 2016

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


On Sat, Oct 15, 2016 at 02:11:11PM +0200, Gandalf Corvotempesta wrote:
> 2016-10-15 13:56 GMT+02:00 Dennis Jacobfeuerborn <dennisml at conversis.de>:
> > Keep in mind that it is never possible to be 100% sure that no
> > split-brain can occur. You might always run into freak accidents.
> 
> Yes, I know. 100% availability is a myth.
> 
> > When you say you use bonded interfaces for redundancy are the physical
> > ports actually connected to separate switches? For proper redundancy the
> > whole network path between the systems must be redundant including any
> > switches that sit in between.
> 
> I'll use direct connections, without switches between drbd nodes.
> 
> Is the 3 node replication fully supported with DRBD 9 ? This should
> remove the split brain risks are 3 nodes won't be split brained
> (3 or any odd number of nodes)

There is no "write quorum" yet, but I'm working on that.

Data divergence is still very much possible.

The DRBD 8.4 integration with pacemaker and fencing mechanisms
is proven to work, whereas the DRBD 9 integration with pacemaker
and fencing mechanisms is still pretty much non-exisitant.

-- 
: Lars Ellenberg
: LINBIT | Keeping the Digital World Running
: DRBD -- Heartbeat -- Corosync -- Pacemaker

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



More information about the drbd-user mailing list