I have a gluster volume which consists of 5 peers (called gl1 - gl5). Only 2 of the peers are bricks, i.e. they store data in replicate/mirror mode; 3 peers only access data from the two bricks. All use glusterfs 3.2.4. Every peer mounts gluster using this /etc/fstab entry: localhost:/shared /shared glusterfs defaults,_netdev 0 0 Unfortunately, what happens, is that all peers, including the brick keeping the data, read files from one brick only: gl1. Is there a way to specify which brick should be preferred for reads? In my opinion, it doesn't make much sense if "gl5" brick reads files from "gl1" brick: "gl5" has these files locally already. Also, I don't like the situation that all other bricks seem to read data only from "gl1", but they almost never read anything from "gl5". root at gl1:~# gluster volume info Volume Name: shared Type: Replicate Status: Started Number of Bricks: 2 Transport-type: tcp Bricks: Brick1: gl1:/data/gluster Brick2: gl5:/data/gluster Options Reconfigured: performance.cache-refresh-timeout: 60 performance.cache-size: 2048MB performance.write-behind-window-size: 128MB performance.io-thread-count: 32 performance.flush-behind: on nfs.disable: on auth.allow: 192.168.*,127.* root at gl1:~# gluster peer status Number of Peers: 4 Hostname: gl2 Uuid: 73cb951a-39f8-4876-82f0-cc9e26462d89 State: Peer in Cluster (Connected) Hostname: gl3 Uuid: 09c1fdd5-cdbc-4174-9dcb-7785546bff5d State: Peer in Cluster (Connected) Hostname: gl4 Uuid: f5bf2828-c673-4d6c-8622-c7586d539ead State: Peer in Cluster (Connected) Hostname: gl5 Uuid: 8c8d980f-15f2-4345-90f2-f75365bf9812 State: Peer in Cluster (Connected) -- Tomasz Chmielewski http://wpkg.org