On Wed, Dec 7, 2016 at 11:16 PM, Steve Taylor <steve.taylor@xxxxxxxxxxxxxxxx> wrote: > > I'm seeing the same behavior with very similar perf top output. One server with 32 OSDs has a load average approaching 800. No excessive memory usage and no iowait at all. Also seeing AVC's so I guess I'll switch to permissive mode for now: type=AVC msg=audit(1481149036.203:819): avc: denied { read } for pid=3401 comm="ceph-osd" name="nvme0n1p13" dev="devtmpfs" ino=18529 scontext=system_u:system_r:ceph_t:s0 tcontext=system_u:object_r:nvme_device_t:s0 tclass=blk_file type=AVC msg=audit(1481149036.203:820): avc: denied { open } for pid=3401 comm="ceph-osd" path="/dev/nvme0n1p13" dev="devtmpfs" ino=18529 scontext=system_u:system_r:ceph_t:s0 tcontext=system_u:object_r:nvme_device_t:s0 tclass=blk_file type=AVC msg=audit(1481149036.203:821): avc: denied { getattr } for pid=3401 comm="ceph-osd" path="/dev/nvme0n1p13" dev="devtmpfs" ino=18529 scontext=system_u:system_r:ceph_t:s0 tcontext=system_u:object_r:nvme_device_t:s0 tclass=blk_file type=AVC msg=audit(1481149036.203:822): avc: denied { ioctl } for pid=3401 comm="ceph-osd" path="/dev/nvme0n1p13" dev="devtmpfs" ino=18529 scontext=system_u:system_r:ceph_t:s0 tcontext=system_u:object_r:nvme_device_t:s0 tclass=blk_file type=AVC msg=audit(1481149036.204:823): avc: denied { write } for pid=3401 comm="ceph-osd" name="nvme0n1p13" dev="devtmpfs" ino=18529 scontext=system_u:system_r:ceph_t:s0 tcontext=system_u:object_r:nvme_device_t:s0 tclass=blk_file I must say I'm a bit surprised that these issues weren't caught in QA. _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com