Hi, I''m using Lustre1.5.91 and kernel 2.6.34(1MDS + 8OST). One of OST was crashe. I have operated it with RAD5. But two disks have broken at the same time . The following log is output. Nov 2 21:16:55 mdgcfs1 kernel: Lustre: 4256:0:(lov_qos.c:403:qos_shrink_lsm()) using fewer stripes for object 0x1782a62: old 8 new 7 It seems temporarily to have operated with seven OST. But it could not access in recovery efforts crashing disk . So I want to restart operation by seven OST piece remaining. What should I do? Junya Shinohara -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.clusterfs.com/pipermail/lustre-discuss/attachments/20061105/d509f927/attachment.html