2011/6/29 Darren Austin <darren-lists at widgit.com>: > > There is no sync between the two servers in the situation I outlined, and the client cannot trigger a self-heal as you suggest because the client is effectively dead in the water until it's forcibly killed and re-mounted. > some information: - you have 2 GlusterFS as server (are "pingable" from any server and from any client ?) - 10.58.139.217 is a GlusterFS server ? ... if so, can you post the "ps -ef | grep gluster" result ? - all GlusterFS version (server and client) are the same ? (glusterfs -v)
