Ben Wang wrote:
Is solution D out of the picture already? I mean if we really
can't
find a good enough solution to solve it, why not just accpet it. State
transfer, IMO, should not happen that often anyway, if it is just a
new node joining. I think the important thing is keep the state
valid/consistent.
It's not about state transfer happening a lot, but when it does happen,
it fails 30% of the time when other members are updating the cluster.
That is a very common use case, unfortunately !
This is a problem that needs to be solved
--
Bela Ban
Lead JGroups / Manager JBoss Clustering Group
JBoss - a division of Red Hat