On Jan 24, 2008, at 11:59 AM, GAUTIER Hervé wrote:
Andrew Beekhof a écrit :
On Jan 24, 2008, at 10:20 AM, GAUTIER Hervé wrote:
Hi there !
As Sylvain, I don't understand how HeartBeat (and in particular
PaceMaker) ensure high availability of a service (application
resource) if it is not possible to configure automatic failover in
case of resource failure needed by the service ?
It _does_ do automatic failover.
What it wont do is move to a node where the resource has _already_
failed.
That's my point.
The rationale is that if something happened on nodeX enough times to
cause the resource migrate away, then there is unlikely to be any
point trying nodeX again until the admin has fixed the underlying
problem (and re-set the failcount).
We want to avoid situations where resources bounce around the cluster
(causing undue churn and load) indefinitely without any chance of
actually running anywhere.
Note also that this is all talking about resource-level failures and
does not apply if the whole node fails.
Soon we will be able to do even better than this by using the new
LRM timing data to expire failures automagically (thus allowing a
resource that fails for a second or third time to return to the
original node after a configured interval)
Ok, thank for the information.
If I understand well, the fail-count counter will be reset
automagically ?
sort-of
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
