Jiann-Ming Su wrote:
> Since I didn't get a response to either of my previous questions about
> multiple ucast directive and hangman, I'll try to explain what we're
> trying to do. Basically, a potentially showstopping constraint is
> geographic redundancy. This makes running a serial cable between the
> two rather difficult.
>
> Each server has two NICs. Each NIC will be connected to a different
> router, so there will be multiple paths between the nodes. We're
> concerned about a "hangman" situation where communication between both
> nodes goes down and the services we're monitoring with Heartbeat come
> up on both nodes.
>
> Is there an easy way to configure 2.0.x without CRM so that in a state
> of confusion, only the primary node stays up? I mean, the haresources
> already require I specify the preferred primary node. So, is there a
> way to key off of this? Or, will we have to write our own ipfail
> script? Thanks for any tips.
In a CRM-based scenario, we are adding a way to get a third site to give
one of the sites quorum - through a quorum server.
You really need a third site for this. But, it doesn't have to be a
full-fledged cluster node.
This code is a few weeks from being complete. Huang Zhen can give you
the details.
--
Alan Robertson <[EMAIL PROTECTED]>
"Openness is the foundation and preservative of friendship... Let me
claim from you at all times your undisguised opinions." - William
Wilberforce
_______________________________________________________
Linux-HA-Dev: [email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/
