Hi Heath, My OSDs do the exact same thing - consume lots of RAM when the cluster is reshuffling OSDs. Try ceph tell osd.* heap release as a cron job. Here's a bug: http://tracker.ceph.com/issues/12681 Chad _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com