Hi list Several months ago I built a configuration based on Gluster+Xen Server. Each Xenserver offers its disk as a brick in a gluster storage and each Xenserver connects to the glusterFS as NFS repository (kill two birds with one stone...) More or less this scenario is working properly but I am watching in a cluster of two Xenservers that I/O wait is nearly 50% with no vm started up and the two bricks syncronized (~300GB) Gluster Version is 3.2.5 compiled to 2.6.32 Gluster Volume configured like the following: Volume Name: ngs Type: Replicate Status: Started Number of Bricks: 2 Transport-type: tcp Bricks: Brick1: xcp-b40:/mnt/1 Brick2: xcp-b30:/mnt/1 Options Reconfigured: performance.read-ahead: on performance.io-cache: on performance.write-behind-window-size: 4MB performance.io-thread-count: 64 performance.cache-size: 134217728 nfs.trusted-sync: on nfs.port: 2049 Can anyone help me to try to get better performance? PS: The rest of Xenservers glusterfs are configured like this, but they aren't as large as this. -------------------------------------------------------------------------------------------- Busca, compara... pero busca en Qu?! Desc?brelo en http://buscar.que.es/ -------------------------------------------------------------------------------------------- Correo enviado desde http://www.ozu.es
