Can you please post the configuration files? It does not sound like a bug though. Unify joins together the storage of several servers in one single volume. So if one server goes down the unified server will become inaccessible. For redundancy you probably want afr. On Fri, Jul 31, 2009 at 10:28, Yi Ling<lingyi.pro@xxxxxxxxx> wrote: > there are 3 servers and a client in my clustered system. there are 2 > exported directories on each server, one for storage,the other for > namespace. volume files for both servers and client are as follow. > > after all servers and client connecting sucessfully, i stop server 1. when i > simply run 'ls' on mount point of the client, my computer (where client is > running ) hang without any response to keyboard and mouse. is that a > failure of a single point ?? > > im gussing it's a bug. anybody could tell why? thank you ~~~ > > _______________________________________________ > Gluster-devel mailing list > Gluster-devel@xxxxxxxxxx > http://lists.nongnu.org/mailman/listinfo/gluster-devel > >
