This is a major issue, IMHO. What I have done is set up a cache store link to elastic and every record written to Ignite goes to elastic too. When I upgrade a cluster to a new version I read everything back into the new cache.
There is not simple way to approach this when you are dealing with caches, I think.... On Sun, Aug 16, 2015 at 11:12 AM, Gianfranco Murador <mura...@apache.org> wrote: > Hi everyone, > I have noticed that the operation of a cluster requires that all nodes have > the same version of Ignite . I think that this rule is consistent , but > how can I approach a possible upgrade version of the cluster without > compromising reloading the entire grid ? Do you also faced this problem or > do you have a procedure for updating that you'd like recommend me? > Regards > -- > Gianfranco Murador > Igniter and Software Engineer. >
