As Deepal points out, in a cluster u need an identical set of nodes.
Thats why at point Chamikara and I suggested loading all nodes from central
repo using URLBasedAxisConfigurator to ensure identical configuration.
Afkam, if there are nodes specific properties, then this application might
not be a good candiate for clustering.

As for Rampart and Sandesha, can these properties be made serializable?
Chamikara what are your thoughts on this?

Regards,

Rajith

On 5/18/07, Afkham Azeez <[EMAIL PROTECTED]> wrote:



On 5/18/07, Deepal Jayasinghe <[EMAIL PROTECTED]> wrote:
>
>
>
> Afkham Azeez wrote:
> > We have a problem when it comes to replicating properties in our
> > clustering implementation.  There are some properties which are
> > specific to a node, specially properties in the ConfigurationContext.
> hmm , as far as I know in a cluster all the replicates should be the
> same (so they need to have the same state) , so having different
> properties means different states.


The thing is, in our case, the ConfigurationContext may hold information
that is Node specific, and which should not be replicated.

--
Thanks
Afkham Azeez

http://www.wso2.org
GPG Fingerprint: 643F C2AF EB78 F886 40C9  B2A2 4AE2 C887 665E 0760

Reply via email to